Glycopeptide Resistance in Gram-positive Bacteria

Transcription

1 ~ CONTINUING EDUCATION Glycopeptide Resistance in Gram-positive Bacteria David C.E. Speller, William A. Lynn+ and Thomas R. Rogers* Readers are invited to use this article as a self-assessment exercise and to update their knowledge. Illustrative case history A 20-year-old female presented in 1992 with nephritis due to systemic lupus erythematosus. Despite plasmapheresis and immunosuppression she became dialysisdependent. Previous complications included failed renal transplantation, recurrent bacterial and viral infections and cyclophosphamide-related common variable immunodeficiency. In late 1994 she was admitted for placement of a femoral venous shunt for haemodialysis access. She was admitted to a renal unit, which had an ongoing outbreak of colonization and infection with glycopeptide-resistant Enterococcus faecium. She had previously been admitted to this ward on numerous occasions, and multiple courses of antibiotics, including cephalosporins, ciprofloxacin, gentamicin and vancomycin, had been given. At the time of admission she was afebrile and generally well and screening swabs showed no colonization with E. faecium. A right femoral vein gortex graft was inserted; the operation was complicated by a small hematoma. Antibiotic prophylaxis was given, with intravenous cefuroxime and metronidazole for 48 hours. On post-operative day 5 she became febrile, but the surgical wound was clean and there was no obvious infective source. The peripheral leukocyte count was 6.4 x 109/1 and the C-reactive protein 159 mg/l (normal: <I0 mg/l). Empiric therapy was commenced with intravenous ceftazidime and vancomycin. Her chest X-ray was clear, and urine and sputum cultures negative. Blood Corresponding author and reprint requests: D.C.E. Speller FRCP: Antibiotic Reference Unit, Central Public Health Laboratory, 61 Colindale Avenue, London NW9 5HT. UK. Tel: Fax: *W.A. Lynn, MRCP and TR. Rogers, FRCPafh: Department of Infectious Diseases and Bacteriology, Royal Postgraduate Medical School, Hammersmith Hospital, Du Cane Road, London W12 ONN, UK cultures subsequently grew E. faecium. A wound swab from the right groin also grew E. faecium with coliforms. The isolates of E. faecium were resistant to ampicillin, vancomycin and teicoplanin (VanA phenotype) and displayed varying sensitivity to gentamicin. Her clinical condition failed to respond to the treatment and so 72 hours later therapy was begun, on a compassionate use basis, with quinupristid dalfopristin (RP 59500), an unlicensed streptogramin antibiotic combination related to pristinamycin, to which the isolates of E. faecium were sensitive. Despite a reduction in her fever and improvement in her general condition she had a persistent inflammatory response, and blood cultures and wound swabs repeatedly yielded resistant E. faecium. Ten days after treatment with quinupristin/dalfopristin was started, the gortex vein graft was surgically removed and dialysis was continued via a central venous catheter. There was prompt clinical improvement with resolution of the bacteremia. Antibiotics were continued for a further week with no subsequent recurrence of fever or bacteremia. Interestingly, at no time before the initial episode of bacteremia had superficial colonisation with resistant E. faecium been detected, despite repeated investigation of swabs from throat and groin. Discussion points Glycopeptide resistance in gram-positive cocci has emerged as a hospital problem in recent years. The multi-resistant enterococcus has been named the nosocomial pathogen of the 1990 s (1). The correct management of such problems has not been established but in the USA the Hospital Infection Control Practices Advisory Committee has recently issued extensive Recommendations (2). The Unit Glycopeptide-resistant enterococci (GRE) have been encountered in renal units and also in hematology, oncology and transplant units. Recently there have been reports of a particular association with intensive care units (3), including childhood intensive care. 54

2 Speller, Lynn, Rogers, Glycopeptide Resistance in Gram-positive Bacteria 55 Admission during an outbreak It does not seem feasible to close units to new admissions because of the occurrence of GRE, although this would be the usual recommendation in the case of other readily transmissible multi-resistant pathogens, such as niethicillin-resistant Staphylococcus aureus (MRSA). Our policy has been to cohort-nurse colonized patients, detected by weekly screening, in one area of the ward, and to admit new patients to a separate area free from colonized patients. Patients with clinical infection, who receive more frequent and closer medical and nursing attention, are cared for in a single room with full isolation precautions - in the manner recommended for both colonized and infected patients in the recent Recommendations (2). Screening Our practice is to screen all patients in affected wards at weekly intervals with culture of throat and groin swabs and, where possible, feces. Direct culture is made on blood agar containing 25 mg/l neoniycin with application of a paper disc containing 30 pg vancomycin. Preliminary enrichment culture in nutrient broth containing 6 mg/l vancomycin increases the yield of positives. Various selective culture systems are reconimended in the literature (2). Role of previous antibiotic administration Data are accumulating to suggest that cephalosporins and quinolones select for gram-positive bacteria resistant to them, such as enterococci and Clostridium dficile. Antibiotic policies in such units need to be flexible and applied judiciously, with frequent review of the rationale for treatment of individual patients. Vancomycin or teicoplanin may be used as part of unit policy or invoked to deal with infection by emerging gram-positive bacteria. Many outbreaks of GRE have followed increased vancomycin use and this may be a significant factor for the colonization of individual patients; the use of glycopeptides requires to be rigorously controlled (2). Prophylaxis for insertion of a vascular graft We advocate narrow-spectrum prophylaxis aimed at S. aureus, with an isoxazolyl penicillin, such as cloxacillin, possibly with gentamicin. In a unit with problems of GRE cephalosporins and glycopeptides should be avoided. Our recommendation would remain the same in a patient (unlike the present case) already known to be colonized by GKE. We should wish at the present time to reserve the few investigational agents active against such strains, such as quinupristin/dalfopristin, for treatment of actual infection, although the susceptibility of the relevant strain to such agents should be determined. Importance of the infected graft In general, it may be impossible to eradicate infection in the presence of a foreign body or a large extravascular focus with antibiotics. Initial experience with quinupristin/dalfopristin bears out this limitation (4). General multiple choice questions on glycopeptide resistance in gram-positive bacteria In each of the numbered questions at least one and up to$ve Ofthe individual entries are correct. The answers are on the next page. 1. Glycopeptide (vancomycin or teicoplanin) resistance has been reported and confirmed in: a. Norardia asteroides b. Pediococcus pentosaceus c. Enteroroccus gallinarum d. Listeria monocytogenes e. Streptococcus pneumoniae. 2. Glycopeptide resistance in Staphylococcus species: a. Vancomycin resistance in Staphylococcus aureus has now been encountered clinically in the USA and Europe b. Teicoplanin resistance is commoner than vancomycin resistance in coagulasenegative staphylococci c. Among coagulase-negative staphylococci, teicoplanin resistance occurs only in Staphylococcus haemolyticus

3 56 Journal of Clinical Microbiology and Infection, Volume 1 Number 1 d. Laboratory detection of resistance by disc tests is made difficult by poor diffusion characteristics e. Laboratory detection of teicoplanin resistance is highly dependent on the medium used. 3. Glycopeptide resistance in enterococci: a. may be associated with production of novel membrane protein enzymes b. may be due to enzymatic destruction of the antibiotic c. may be due to reduced binding of antibiotic to side-chains of cell wall precursors d. may be mediated by genes carried on transposons e. to teicoplanin is always associated with the vana gene rather than the vanb gene. 4. Glycopeptide-resistant enterococci have been detected and reported a. in farm animals b. to persist for >3 months in hospital patients c. in the feces of hospital staff d. in operating theatre air e. to persist for >3 days in the hospital environment. 5. Characteristic clinical associations reported: a. Vancomycin-resistant enterococci - endocarditis on natural heart valves b. Vancomycin-resistant enterococci - persistent bacteria without endocardltis or extravascular focus C. Vancomycin-resistant Leuconostoc species - bacteremia associated with colonization of intravascular cannula d. Vancomycin-resistant lactobacillus endocarditis - relapse after treatment with penicillin alone e. Teicoplanin-resistant staphylococci - results of treatment poorly correlated with in vitro susceptibility test results. 6. Ampicillin treatment of infections by vancomycin-resistant enterococci: a. High-level (MIC >2000 mg/l) resistance to gentamicin and streptomycin implies absence of synergism with all available aminoglycosides b. Ampicillin alone in high doses may succeed in endocarditis c. Ampicillin resistance may occur by p-lactamase production d. Addition of sulbactam restores sensitivity to all ampicillin-resistant strains e. Ampicillin and vancomycin may show synergism against ampicillin- and vancomycin-resistant strains. 7. Developments in the following antimicrobial groups show promise for the future treatment of infections by multi-resistant enterococci: a. glycopeptides b. cephalosporins c. aminoglycosides d. pristinamycins e. quinolones.. Commentary on multiple choice questions Question 1 Resistance to vancomycin, in the gram-positive species susceptible to it, first appeared among enterococci in the 1980s, when a nosocomial outbreak of infection by vancomycin-resistant enterococci was reported (5). Since that time sporadic cases and clusters have been reported from Europe and the USA. High- and lowlevel resistance has been encountered in Enterococcus Answers to multiple choice questions QZ a. true; b. true; c. true; d. false; e. false 42 a. false; b. true; c. false; d. true; e. true 43 a. true; b. false; c. true; d. true; e. false 44 a. true; b. true; c. true; d. false; e. true 45 a. true; b. true; c. true; d. true; e. true Q6 a. true; b. true; c. true; d. false; e. true 47 a. true; b. false; c. false; d. true; e. true

4 Speller, Lynn, Rogers, Glycopeptide Resistance in Gram-positive Bacteria 57 faecium and Enterococcus faecalis, while low-level constitutive resistance is a characteristic of Enterococcus gallinarum and other species that are less frequently encountered in human infection (6). Resistance to glycopeptides is common in lactic acid bacteria, such as Pediococcus, Leuconostoc and Lactobacillus species (7). These have been noted more frequently as opportunistic pathogens, particularly following glycopeptide administration. In the case of other gram-positive bacteria that are often resistant to glycopeptides, such as Nocardia spp. and Erysipelothrix rhusiopathiae, administration of glycopeptides does not usually arise, either as a predisposing factor or as a therapeutic option. A fear is that the genetic material mediating resistance to the glycopeptides may be transferred to species which are sensitive at present, such as Streptococcus pneumoniae. Question 2 It is particularly worrying that vancomycin resistance can be transferred (in the laboratory) from an enterococcus to Staphylococcus aureus (8), but thus far no confirmed report of resistance in clinical S. aureus isolates has been published. Low-level resistance to teicoplanin, however, has been reported (9). In coagulase-negative staphylococci teicoplanin resistance dominates (lo), although vancomycin resistance has been encountered (11). The earlier reports of high-level teicoplanin resistance concerned Staphylococcus haernolyticus, but this phenomenon is also seen in Staphylococcus epiderrnidis, Staphylococcus warneri, Staphylococcus horninis and Staphylococcus xylosus. Exercises comparing the performance of laboratories in the detection of glycopeptide resistance, in Europe and in the USA, have shown success in the detection of high-level vancomycin resistance in enterococci, but poor results with strains showing resistance at a lower level (12, 13). Several factors contribute to inconsistent laboratory reporting. There are discrepancies between the interpretive breakpoints recommended in different countries (1 4, 15). The diffusion characteristics of the large molecules make the zones of inhibition in disc tests very small and less easy to interpret. In addition, very different MIC values of teicoplanin may be obtained with various media, with and without the addition of blood, and with changes in inoculum (1 6). Question 3 There is no evidence of destruction of glycopeptides by resistant strains of E. faecium and E. faecalis (17). Such strains have been shown to synthesize novel membrane proteins, which in some cases have been shown to possess D-ala-D-ala ligase activity and to produce different peptide side-chains on the cell wall components. These will not bind glycopeptides and, even in their presence, can link up to give the strengthening cross-linking structure to the bacterial cell wall (18). The resistant enterococci first described showed inducible high-level resistance to vancomycin and also resistance to teicoplanin: the VanA phenotype. Other strains were encountered, however, showing lower levels of inducible vancomycin resistance and sensitivity to teicoplanin: the VanB phenotype ( VanC describes the low-level constitutive resistance to vancomycin with sensitivity to teicoplanin seen in E. gallinarum and some other species). The corresponding genes have been identified. More recently strains possessing the vanb gene, rather than the vana gene, have been described with high-level vancomycin resistance and resistance to teicoplanin (1 9). VanA glycopeptide resistance was early demonstrated to be plasmid-borne and transferable. Mobility of this genetic material is enhanced by its occurrence as a transposon. The resistance transposon Tn I546 carries a complex of genes encoding glycopeptide resistance (including vana) and its regulation (20). Question 4 Glycopeptide-resistant enterococci were first encountered in hospitals, and it is believed that antibiotic pressure, first of cephalosporins (to which enterococci are intrinsically resistant) and subsequently of glycopeptides, is important in their emergence. Nevertheless, glycopeptide-resistant enterococci have been discovered in the gastro-intestinal tracts of patients in the community (21), in farm animals and in poultry for sale (22). The carriage rate of patients in intensive care, renal and hematology units may be high, and the glycopeptide-resistant enterococci may persist for long periods (23). The factors affecting patient-to-patient spread during outbreaks are still obscure. Staff may become colonized (24) but this is not a consistent finding. Likewise, contamination of the immediate environment of patients and transient contamination of staff hands have appeared to be significant factors in some outbreaks (25, 26) but not in others. Question 5 Most of the species of glycopeptide-resistant grampositive bacteria under discussion characteristically act as opportunistic pathogens in impaired hospital patients. Even when they are isolated from deep specimens their clinical significance may be difficult to assess. Bacteremia may persist without a defined focus and with seemingly little deleterious effect on the

5 58 Journal of Clinical Microbiology and infection, Volume 1 Number 1 patient (27). Removal of an intravascular cannula may be sufficient to cure the bacteraemia (28). Assessment of clinical results assisted the National Committee for Clinical Laboratory Standards in establishing breakpoints for teicoplanin (1 4), but many investigators have commented on the poor correlation between in vitro results and clinical outcome (29). Endocarditis provides a clear test of the power of an antibiotic regimen. In the case of endocarditis caused by the lactic acid bacteria, definitive advice on treatment cannot be given, but it is clear that a bactericidal synergistic combination should be used (30). Question 6 Unfortunately, glycopeptide-resistant strains of E. faecium are often resistant to 6-lactams and, at highlevel, to aminoglycosides. Gentamicin is the aminoglycoside usually tested. If there is high-level resistance to this agent, it is worthwhile to test streptomycin. High-level resistance to both implies that no clinically available aminoglycoside will give satisfactory synergy (31). In these circumstances, apparent sensitivity to amikacin does not denote sufficient activity to give synergy. If an aminoglycoside-resistant strain is ampicillin-sensitive this agent alone in high doses is the current recommendation in endocarditis (31). Sometimes it is possible to demonstrate, in strains resistant to both p-lactams and glycopeptides, a synergistic activity of a combination of these agents, but this is not a consistent finding (32). Resistance caused by p-lactamase production has been described, especially in the USA, and the P-lactamase is susceptible to clavulanate and to sulbactam (33), but most p-lactam-resistant strains are resistant by a different mechanism: modification of penicillin-binding proteins. Question 7 Because of these increasing therapeutic problems, there is a need to search for new active compounds. Investigation of p-lactams and aminoglycosides has not been fruitful. The clinical activity of newer peptides, such as daptomycin, has been disappointing thus far, although further compounds with much greater activity are under development and appear promising (34, 35). New fluoroquinolones, with increased activity against gram-positive bacteria, are being developed (36) and offer some promise although they require clinical trial in these infections. The injectable pristinamycin combination, quinupristin/ dalfopristin has been used with success against some infections by multi-resistant enterococci (37). This is more active against E.faecium than against E.faecalis and appears to have almost no bactericidal activity against enterococci. Bacteriostatic activity is also shown by the glycylcyclines (tetracycline derivatives) (38). Despite the comparatively non-pathogenic character of many of the species that show glycopeptide resistance, the need to treat infections caused by them is becoming more frequent. The ready transmissibility of the underlying genetic material poses a serious threat to our future use of antibiotics. Acknowledgements We thank Dr Elaine Clutterbuck for permission to report the case, and we are most grateful to our colleagues, Dr A.P. Johnson, Dr N. Woodford and Mr D. Morrison, for their assistance in the production of this article. We thank RhBne-Poulenc Rorer, Paris for providing QuinopristidDalfopristin for compassionate use. This article is intended to inaugurate a series of exercises for Continuing Education in the disciplines of Clinical Microbiology and Infectious Diseases. The Editors will be interested to receive comments on the relevance and usejulness ofthis article and on its format, and also suggestionsfor-future topics. Further reading Woodford N, Johnson AP, Morrison D, Speller DCE. Current perspectives of glycopeptide resistance. Clin Microbiol Rev 1995; in press. References 1. Spera RV, Farber BE Multiply resistant Enterococcusfaecium. The nosocornial pathogen of the 1990's. J Amer Med Assoc 1992; 268: Hospital Infection Control Practices Advisory Committee. Recommendations for preventing the spread of vancomycin resistance. Inf. Control Hosp. Epidemiol. 1995; 16: Centers for Disease Control and Prevention. Nosocomial enterococci resistant to vancomycin - United States, MMWK 1993; 42: Linden P, Pasculle AW, Kidder S, Kusne S, Silverman A, Dean J. Quinupristin/dalfopristin (RP 59500) for the treatment of serious infection due to high level vancomycin resistant Enterococcus faecium. Program Abstr 34th Intersci ConfAntimicrob Agents Chemother 1994; session 126: M Uttley AHC, George KC, Naidoo J et al. High-level vancomycin-resistant enterococci causing hospital infection. Epid Infect 1989; 103: Leclercq R, Dutka-Malen S, Duval J, Courvalin l? Vancomycin resistance gene vunc is specific to Enterocomrs gdinarum. Antimicrob Agents Chemother 1992; 36: Swenson JM, Facklam RR, Thornsberry C. Antimicrobial susceptibility of vancomycin-resistant Leuconosioc, Pediococcus

6 Speller, Lynn, Rogers, Glycopeptide Resistance in Gram-positive Bacteria 59 and Lactobarillus species. Antimicrob Agents Chemother 1990; 34: Noble WC, Virani Z, Cree RGA. Co-transfer of vancomycin and other resistance genes from Enferocorcus Jiecalis NCTC to Staphylococcus airreus. FEMS Microbiol Lett 1992; 93: Vedel G, Leruez M, Lemann F, Hraoui E, Katovahery D. Prevalence of Staphylococcus aureus and coagulase-negative staphylococci with decreased sensitivity to glycopeptides as assessed by determination of MICs. Eur J Clin Microbiol Infect Dis 1990; 9: Goldstein FW, Coutrot A, Sieffer A, Acar JE Percentages and distributions of teicoplanin- and vancomycin-resistant strains among coagulase-negative staphylococci. Antimicrob Agents Cheniother 1990; 34: , 11. Veach LA, Pfaller MA, Barrett M, Koontz FP, Wenzel RP. Vancomycin resistance in Staphylococcus haemolyticus causing colonization and bloodstream infection. J Clin Microbiol 1990; 28: Snell JJS, Brown DFJ, Perry SF, George R. Antimicrobial susceptibility testing of enterococci: results of a survey conducted by the United Kingdom National External Quality Assessment Scheme for Microbiology. J Antimicrob Cheniother 1993; 32: Tenover FC, Tokars J, Swenson J, Paul S, Spitalny K, Jarvis W. Ability of clinical laboratories to detect antimicrobial agent-resistant enterococci. J Clin Microbiol 1993; 31: National Committee for Clinical Laboratory Standards. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically - third edition. Approved Standard M7-A National Committee for Clinical LabOrdtOry Standards, Villanova, Pa. 15. Working Party of the British Society for Antimicrobial Chemotherapy. A guide to sensitivity testing. J Antimicrob Chetnother 1991; 27 (Suppl. D): Felmingham D, Solomonides K, O'Hare MD, Wilson APR, Griineberg RN. The effect ofmedium and inoculum on the activity of vanconiycin and teicoplanin against coagulasenegative staphylococci. J Antimicrob Chemother 1987; 20: Johnson AP, Uttley AHC, Woodford N, George KC. Resistance to vancomycin and teicoplanin: an emerging clinical problem. Clin Microbiol Rev 1990; 3: Woodford N, Johnson A. Glycopeptide resistance in grampositive bacteria; from black and white to shades of grey J Med Microbiol 1994; 40: Hayden MK, Trenholme GM, Schultz JE, Sahm DF. In vivo development of teicoplanin resistance in a VanB Enterococcus faecium isolate. J Infect Dis 1993; 167: Arthur M, Molinas C, Depardieu F, Courvalin I? Characterization of Tn 1546, a Tn3-related transposon conferring glycopeptide resistance by synthesis of depsipeptide peptidoglycan precursors in Enferococctrsfaecium BM4147. J Bacteriol 1993; 175: Jordens JZ, Bates J, Griffiths DT. Faecal carriage and nosocomial spread of vancomycin-resistant Enteroroccus faecium. J Antimicrob Chemother 1994; 34: Bates J, Jordens JZ, Griffiths DT. Farm animals as a putative reservoir for vancomycin-resistant enterococcal infection in man. J Antimicrob Chemother 1994; 34: Boyce JM, Opal SM, Chow JW et al. Outbreak of multidrug-resistant Enterococcus faecium with transferable uanb class vancomycin resistance. J Clin Microbiol 1994; 32: Handwerger S, Raucher B, Altarac D et al. Nosocomial outbreak due to Enterococcus faecium highly resistant to vancomycin, penicillin and gentamicin. Clin Infect Dis 1993; 16: Karanfil LV, Murphy M, Josephson A et al. A cluster of vancotnycin-resistant Enterococcus faaeciim in an intensive care unit. Infect Control Hosp Epidemiol 1992; 13: Rhinehart E, Smith N, Wennersten C et al. Rapid dissemination of beta-lactamase-producing aminoglycosideresistant Enterococcus faecium. N Engl J Med 1990; 323: Gray J, March PJ, Stewart D, Pedler SJ. Enterococcal bacteraemia: a prospective study of 125 episodes. J Hosp Infect 1994; 27: Handwerger S, Horowitz H, Coburn K, Kolokathis A, Wormser GP. Infection due to kuconostoc species: six cases and review. Rev Infect Dis 1990; 12: O'Connell B, Browne PV, Cafferkey MT, McCann SK. Coagulase-negative staphylococcal bacteraemia treated with teicoplanin. J Antimicrob Chemother 1993; 31: Griffiths JK, Daly JS, Dodge RA. Two case5 of endocarditis due to Lactobacillus species: antimicrobial susceptibility, review and dxcussion oftherapy. Clin Infect Dis 1992; 15: 25@255. Eliopoulos GM. Aminoglycoside resistant enterococcal endocarditis. Infect Dis Clin N Amer 1993; 7: Gutmann L, Al-Obeid S, Billot-Klein D, Guerrier M-L, Collatn E. Synergy and resistance to synergy between p- lactam antibiotics and glycopeptides against glycopeptideresistant strains of Enterococcus faecium. Antimicrob Agents Cheniother 1994; 38: Lavoie SR, Wong ES, Coudron PE, Williams DS, Markowitz SM. Comparison of ampicillin-sulbactam with vancomycin for treatment of experimental endocarditis due to a p-lactamase-producing, highly gentamicin-resistant isolate of Enterococcusfaecalis. Antimicrob Agents Cheniother 1993; 37: Felminghatn D. Towards the ideal glycopeptide. J Antimicrob Chemother 1993; 32: Nicas TI. Activity of novel semi-synthetic glycopeptides against vancomycin-resistant enterococci. Program Abstr 7th Eur Cong Clin Microbiol Infect Dis 1995: abstr 560. Piddock LJV. New quinolones and gram-positive bacteria. Antimicrob Agents Chemother 1994; 38: Lynn WA, Clutterbuck E, Want S et al. Treatment of CAPD-peritonitis due to glycopeptide-resistant Enterococcus faecium with quinupristin/dalfopristin. Lancet 1994; 344: Eliopoulos GM, Wennersten CB, Cole G, Moellering RC. In vitro activity of two glycylcyclines against grampositive bacteria. Antiniicrob Agents Chemother 1994; 38: