MedicineLakex

 

Dkf009.fm

Journal of Antimicrobial Chemotherapy (2002) 49, 999–1005
DOI: 10.1093/jac/dkf009
Risk factors associated with nosocomial methicillin-resistant
Staphylococcus aureus (MRSA) infection including previous
use of antimicrobials
Eileen M. Graffunder* and Richard A. Venezia
Department of Epidemiology MC-45, Albany Medical Center Hospital, 43 New Scotland Avenue, Albany, NY 12208, USA Received 9 May 2001; returned 31 October 2001; revised 10 December 2001; accepted 25 January 2002 Methicillin-resistant Staphylococcus aureus (MRSA) is a major nosocomial pathogen world-
wide. To investigate an association between antimicrobial use and MRSA, a case control study
of 121 patients infected with MRSA compared with 123 patients infected with methicillin-
susceptible S. aureus (MSSA) was carried out. Antimicrobial use was analysed by three different
logistic regression models: all β-lactam antibiotics, β-lactam antibiotics grouped in classes and
antimicrobial use in grammes. Patients infected with MRSA tended to have more co-morbidities,
longer lengths of stay (LOS) and greater exposure to antibiotics than MSSA-infected patients.
Multivariate analysis identified levofloxacin [odds ratio (OR) 8.01], macrolides (OR 4.06),
previous hospitalization (OR 1.95), enteral feedings (OR 2.55), surgery (OR 2.24) and LOS before
culture (OR 1.03) as independently associated with MRSA infection. All models were concordant
with the exception of macrolides, which were not significant based on the number of grammes
administered. There were no significant differences in the types of infection or the attributed
mortality in either group. MRSA-infected patients had a significantly longer LOS before infection
[18.8 ± 18.2 compared with 8.4 ± 6.9 (P < 0.001)] and a significantly longer post-diagnosis LOS
[27.8 ± 32.9 compared with 18.6 ± 21 (P = 0.01)] than MSSA-infected patients.
The reasons for the emergence of MRSA are multifactorial and can be attributed to host factors, infection control prac- Methicillin-resistant Staphylococcus aureus (MRSA) is a tices and antimicrobial pressures.12–14 The appearance of major nosocomial pathogen worldwide.1–3 The Centers for bacterial resistance phenotypes has been linked to the clinical Disease Control (CDC) National Nosocomial Infection use of antimicrobial agents to which the bacteria express Surveillance System (NNIS) reported that methicillin resist- resistance.15 One study demonstrated that a patient's normal ance among S. aureus in US hospitals increased from 2.4% in colonizing flora changes within 24–48 h under selective anti- 1975 to 29% in 1991, with a higher degree of resistance in in- biotic pressures.16 A recent review of more than 20 studies byMonnet & Frimodt-Moller17 identified consistent associ- tensive care units.4–5 More recent data from 1990 through ations and dose–effect relationships that support a causal rela- 1997 identified that the MRSA incidence rate increased 260% tionship between MRSA and antimicrobial drug use. One of in hospitals that participated in the International Networks for these studies demonstrated that ciprofloxacin and cephalo- the Study and Prevention of Emerging Antimicrobial Resist- sporins promoted the colonization and ultimately the spread ance (INSPEAR) Programme.6 Traditionally, MRSA was of MRSA in one hospital.18 In studies where antimicrobial identified infrequently from patients in the community, but classes are analysed separately both cephalosporins and over the last few years reports have documented increases in fluoroquinolones are often identified as risk factors for community MRSA, which may suggest a changing epidemi- MRSA.19 A multi-centre study of 50 Belgian hospitals asso- ciated an increasing incidence of MRSA with increasing use *Corresponding author. Tel: +1-518-262-8230; Fax: +1-518-262-3745; E-mail: [email protected] 2002 The British Society for Antimicrobial Chemotherapy E. M. Graffunder and R. A. Venezia
Table 1. Patient characteristics (before MRSA/MSSA infection) that were significantly different in the MRSA-infected
population
MRSA (n = 121) MSSA (n =123) LOS before culturea Unit changesa Total number of grammes of antibioticsa Central line daysa Urinary catheter daysa Ventilator daysa Respiratory therapy daysa Previous hospitalization Rehabilitation unit ET tube/Trach/NG tubeb Urinary catheters Previous antibiotics Total parenteral nutrition Enteral feedings within 2 h of oral levofloxacin dose 17.3 (2.3, 127.8) Abdominal surgery Multi-organ failure Skin ulcers, eczema, etc.
aMean and range.
bET, endotracheal; Trach, tracheostomy; NG, nasogastric.
of ceftazidime and cefsulodin, co-amoxiclav and fluoro- Materials and methods
quinolones.19 Dziekan et al.20 also showed that fluoroquino-lone use was an independent risk factor for MRSA as well as This study was conducted at Albany Medical Center, a 600 nasogastric tubes and central venous catheters. In separate bed tertiary care facility. Cases were patients infected with studies, both Crossley et al.21 and Hershow et al.22 noted that nosocomial MRSA from 1997 to 1999. Controls were se-lected randomly from a database of all patients infected with patients with MRSA infection had a significantly longer nosocomial MSSA from the same time period. A nosocomial length of stay (LOS) before infection and were likely to have case was defined as a patient without any evidence of infec- received antimicrobial therapy. A recent publication identi- tion on admission and who was culture positive >48 h after ad- fied levofloxacin therapy, ICU setting and LOS as being mission. A retrospective review of the patient's medical independently associated with MRSA.23 record was conducted using a standardized data collection Over the past 3 years, there has been a significant increase form. All of the data were collected before MRSA or MSSA in the incidence of nosocomial MRSA at our institution. Dur- infection. Antibiotic data were grouped into antimicrobial ing this time, there was no significant difference in the classes for analysis. Infections were defined using the CDC number of admissions or patient days. Concurrent with this NNIS definitions.24 Crude and attributable mortality rates increase in MRSA incidence was an increase in the usage of were calculated. Death was considered attributed to infection fluoroquinolone and β-lactam/β-lactamase inhibitor anti- if the infection either directly caused death or exacerbated a microbials. The purpose of this study was to identify pre-existing condition that otherwise would not have resulted characteristics that were associated with nosocomial MRSA in the patient's death. Antimicrobial susceptibility testing on S. aureus was carried out by disc diffusion according to Risk factors associated with nosocomical MRSA
Table 2. Patient characteristics (before MRSA/MSSA infection) not significantly different in the MRSA-
infected population
MRSA (n = 121) MSSA (n = 123) Chest tube daysa Prosthetic devices Haemodynamic instability H blockers/ion pump Radiation therapy (previous year) Respiratory therapy History of aspiration aMean and range.
bICU, intensive care unit; HIV, human immunodeficiency virus; ARDS, adult respiratory distress syndrome; COPD, chronic obstructivepulmonary disease.
National Committee for Clinical Laboratory Standards (NC- period. Tables 1 and 2 list the ORs and significant levels for CLS).25 This study was approved by the institutional review potential risk factors collected. MRSA-infected patients tended to have longer LOS before infection, and had moreward changes and device days. With regard to hospital loca- tion, there was a significant difference in the two units. Boththe medical intensive care and rehabilitation units had a Statistical analysis was carried out using SPSS Version 8 significantly higher number of patients infected with MRSA.
(SPSS, Chicago, IL, USA). Continuous variables were com- Patients infected with MRSA tended to have more antibiotics, pared with a t-test for independent samples (two-tailed). Di- particularly β-lactam antibiotics, levofloxacin and macro- chotomous variables were compared with a two-tailed lides (Table 3).
Fisher's exact test for 2 × 2 comparisons or a Pearson's χ2 test Multiple logistic regression analysis identified six risk for greater than two variables. Odds ratios (ORs) and their factors that were associated independently with MRSA 95% confidence intervals (CIs) were computed. A logistic infection. These risk factors were previous hospitalization, regression model was carried out by a forward selection using longer LOS before infection, surgery, enteral feedings, levo- the likelihood ratio statistic.
floxacin use and macrolide use. Three logistic regressionmodels were carried out to assess antimicrobial usage. The first model included the significant β-lactam antibioticsgrouped in classes, whereas the second model combined all Univariate analysis was conducted on 121 MRSA-infected the β-lactam antibiotics together. This was conducted because patients and 123 MSSA-infected patients from the same time the selective pressures for all β-lactam antibiotics are approxi- E. M. Graffunder and R. A. Venezia
Table 3. Univariate analysis of antimicrobial therapy before MRSA/MSSA infection
Antimicrobial class MRSA (%) (n = 121) MSSA (%) (n = 123) β-Lactam/β-lactamase inhibitor combinations 1st generation cephalosporins 2nd generation cephalosporins 3rd generation cephalosporins All β-lactam antibiotics mately the same, and segregating or combining them may infected population, 12.4% versus 5.7% ( P = 0.08), and soft have diluted their effect. The third model substituted the tissue infections were higher in the MSSA-infected popula- number of grammes of drug administered for significant anti- tion, 14.6% versus 9.1% (P = 0.24). Less common infections microbial agents to assess for dose–effect relationships. The included urinary tract infections, 5.0% and 4.1% (P = 0.77); results of the three models were concordant, with the excep- cardiovascular system infections, 4.1% and 4.9% (P = 1.0); tion of macrolides, which were not significant based on the infections of the ears, eyes and oral cavity, 1.7% and 2.4% number of grammes (Tables 4 and 5).
(P = 1.0); and infections of the reproductive system, 0.8% and With regard to outcome, the only significant difference be- 0.8% (P = 1.0), for MRSA- and MSSA-infected patients, tween MRSA-infected and MSSA-infected patients was respectively. The remaining three infections were gastro- LOS. The mean LOS for MRSA-infected patients was intestinal (one), and bone and joint system (two).
significantly longer (46.1 ± 38.1 compared with 26.2 ± 20.7, Although patients infected with MRSA tended to have a P < 0.001) than MSSA-infected patients. MRSA-infected higher crude mortality rate, 28.9% compared with 19.5% patients had longer LOS before infection (18.8 ± 18.2 (P = 0.10), most of these deaths were not related to infection.
versus 8.4 ± 6.9, P < 0.001), and longer post-diagnosis LOS In a separate analysis, independent risk factors for attributed (27.8 ± 32.9 versus 18.6 ± 21, P = 0.01). There was no differ- mortality were bacteraemia with an OR of 4.2 (95% CIs 1.3, ence in the types of infection identified. The most common 13.5), diabetes OR 3.9 (95% CIs 1.2, 12.0), kidney failure OR types of infection were bloodstream infections, 23.1% and 4.9 (95% CIs 1.4, 16.5), steroids OR 3.5 (95% CIs 1.0, 12.1) 29.3% (P = 0.31); lower respiratory tract infections, 21.5% and haemodynamic instability OR 5.7 (95% CIs 1.8, 18.7).
and 20.3% (P = 0.88); and pneumonia, 21.5% and 15.4% The attributed mortality rate for MRSA-infected patients, (P = 0.25), for MRSA- and MSSA-infected patients, respect- 8.3%, was not significantly different from that of the MSSA- ively. Surgical site infections were higher in the MRSA- infected patients, 5.7% (P = 0.46).
Table 4. Multiple logistic regression analysis of factors
Table 5. Multiple logistic regression analysis of factors
associated with MRSA infection (models I and II) associated with MRSA infection (model III) Number of grammes of Previous hospitalization Previous hospitalization LOS before culture LOS before culture Risk factors associated with nosocomical MRSA
S. aureus resistance.44,45 Isaacs et al.44 postulated that thefactors which predispose S. aureus to develop methicillin Factors that were independently associated with MRSA in- resistance may also predispose them to ciprofloxacin resist- fection were previous hospitalization (within the last 12 ance. In their study, they identified an increase in cipro- months), longer LOS before infection, previous surgery, en- floxacin resistance in isolates after using ciprofloxacin to treat teral feedings, macrolide use and levofloxacin use. Based on MRSA infections. Although there are no known biological the results of the three models the role of macrolides is debat- mechanisms for fluoroquinolones to select resistance to β- able. All of the other risk factors were concordant across the lactam antibiotics in staphylococci, evidence is mounting that three analyses with the most significant risk factors being fluoroquinolones exhibit some type of influence on MRSA.
enteral feedings and levofloxacin use. Previous hospitaliza- Ciprofloxacin resistance in MRSA developed rapidly tion and longer LOS before infection are well known risk whereas the rate of ciprofloxacin resistance in MSSA nation- factors for antimicrobial resistance.26–28 These factors may wide is c. 2.4%. These data indicate that there are intrinsic represent chronic illness and previous exposure to antibiotics factors which lead to greater acquisition of fluoroquinolone as well as opportunities to be colonized with resistant organ- resistance in MRSA.46 Hetero-resistant S. aureus populations isms. Patients with MRSA infections tended to have more co- include sub-populations that contain the mecA gene and these morbidities but these factors failed to achieve statistical resistant sub-populations can be selected for by exposure to significance in multivariate analysis. Surgery has previously increasing concentrations of antibiotics below MIC levels.16 been identified as a risk factor for MRSA infection and may It was demonstrated that mecA-positive S. aureus strains represent a breakdown of the normal host defences, surgical which exhibit this hetero-resistance showed an increase in the technique or post-operative care.29 Surgical site infections proportion of oxacillin-resistant cells following exposure to were 50% more prominent in the MRSA-infected population.
Enteral feedings may represent greater severity of illness in In summary, patients infected with MRSA were more the MRSA-infected population or may have served as a portal likely to have had surgery, a previous hospitalization and a of entry for MRSA. Contamination of enteral nutrition solu- longer LOS before infections. All of these factors are known tions can occur during the assembly or administration via a to increase the probability of a patient developing an MRSA contaminated feeding tube. Factors that contribute to the infection. The two risk factors that contributed the greatest contamination of enteral feedings include the duration of risk for developing an MRSA infection were enteral feedings administration, the composition of the enteral solution and the and levofloxacin as shown by their ORs and highly significant number of manipulations in the feeding process. MRSA- P values. Whether these two risk factors have an associated infected patients tended to have received more concurrent effect is unknown and beyond the scope of this analysis.
dosing of oral levofloxacin and enteral feeds. Previous studies Further investigation is warranted because this could lead to a have demonstrated that a significant reduction in fluoro- modifiable practice change. The relationships between anti- quinolone bioavailability (26–72%) occurs when these drugs microbial use and MRSA are complex and more studies that are co-administered with enteral feedings.30–34 Clinically, this address these issues are needed. However, most studies agree may contribute to therapeutic failure but there have been that, in addition to good infection control practices, the limited studies conducted that directly assess outcome or prudent use of antimicrobial agents is one of the major steps to development of antibiotic resistance as a result of decreased reducing the growing problem of antibiotic resistance.
bioavailability. Further studies designed directly to assess thisrelationship need to be conducted.
Patients who received levofloxacin had the highest risk for MRSA infection with an OR of 8.01. More MRSA-infected 1. Wenzel, R. P., Nettleman, M. D., Jones, R. N. & Pfaller, M. A.
patients received macrolides but macrolides did not retain (1991). Methicillin-resistant Staphylococcus aureus: implications for statistical significance based on the total amount of drug the 1990s and effective control measures. American Journal of administered. In our adult population from 1997 to 1999, Medicine 91, 221S–7S.
33.9% of the S. aureus isolates were resistant to levofloxacin 2. Witte, W. (1999). Antibiotic resistance in Gram positive
and 44.8% were resistant to macrolides. Antimicrobial use bacteria: epidemiological aspects. Journal of Antimicrobial Chemo- has historically been associated with MRSA and a number of therapy 44, 1–9.
different classes of antibiotics.12–23,27,28,35–41 Mechanisms of 3. Cunha, B. A. (1998). Antibiotic resistance. Control strategies.
fluoroquinolone resistance in S. aureus, DNA gyrase, IV Critical Care Clinics 14, 309–27.
topoisomerase and NorA-mediated efflux and macrolide 4. Panlilio, A. L., Culver, D. H., Gaynes, R. P., Banerjee, S., Hender-
resistance ermA and ermC genes are well documented.40,42,43 son, T., Tolson, J. et al. (1992). Methicillin-resistant Staphylococcus In addition to the molecular studies, several clinical studies aureus in U. S. hospitals, 1975–1991. Infection Control and Hospital have shown an association with fluoroquinolone use and Epidemiology 13, 582–6.
E. M. Graffunder and R. A. Venezia
5. Archibald, L., Phillips, L., Monnet, D., McGowan, J. E., Tenover,
20. Dziekan, G., Hahn, A., Thune, K., Schwarzer, G., Schafer, K.,
F. & Gaynes, R. (1997). Antimicrobial resistance in isolates from Daschner, F. D. et al. (2000). Methicillin-resistant Staphylococcus inpatients and outpatients in the United States: increasing import- aureus in a teaching hospital: investigation of nosocomial trans- ance of the intensive care unit. Clinical Infectious Diseases 24,
mission using a matched case-control study. Journal of Hospital Infection 46, 263–70.
6. Bartlett, M., McIntyre, L., Moseley, R., O'Connor, J., Shaw, J. &
21. Crossley, K., Loesch, D., Landesman, B., Mead, K., Chern, M.
Whitehead, C. (2000). Methicillin-resistant S. aureus rates increase & Strate, R. ( 1979). An outbreak of infections caused by strains of 260% in hospitals participating in HIP's international monitoring Staphylococcus aureus resistant to methicillin and aminoglycosides systems. Centers for Disease Control and National Center for I: clinical studies. Journal of Infectious Diseases 139, 273–9.
Infectious Diseases Focus 9, 4.
22. Hershow, R. C., Khayr, W. F. & Smith, N. L. (1992). A compar-
7. Akram, J. & Glatt, A. E. (1998). True community-acquired
ison of clinical virulence of nosocomially acquired methicillin resist- methicillin-resistant Staphylococcus aureus bacteremia. Infection ant and methicillin-sensitive Staphylococcus aureus infections in a Control and Hospital Epidemiology 19, 106–7.
university hospital. Infection Control and Hospital Epidemiology 13,
8. Centers For Disease Control. (1999). Four pediatric deaths
from community-acquired methicillin-resistant Staphylococcus 23. Fu-Yu, C., Muddoloro, M. & Alcid, D. (2000). Levofloxacin
aureus—Minnesota and North Dakota, 1997–1999. Morbidity and associated increase in methicillin-resistant Staphylococcus aureus Mortality Weekly Report 48, 707–10.
infections. In Program and Abstracts of the Thirty-eighth Annual 9. Goetz, A., Posey, K., Fleming, J., Jacobs, S., Boody, L., Wage-
Meeting of the Infectious Diseases Society of America, New ner, M. et al. (1999). Methicillin-resistant Staphylococcus aureus in Orleans, LA, 2000. Abstract 163, p. 241. Infectious Diseases the community: a hospital-based study. Infection Control and Hospital Society of America, New Orleans, LA.
Epidemiology 20, 689–91.
24. Garner, J. S., Jarvis, W. R., Emori, T. G., Horan, T. C. &
10. Kallen, A., Driscoll, T., Thorton, S., Olson, P. & Wallace, M.
Hughes, J. M. (1988). CDC definitions for nosocomial infections.
(2000). Increase in community-acquired methicillin-resistant American Journal of Infection Control 16, 128–40.
staphylococcus at a naval medical center. Infection Control and 25. National Committee for Clinical Laboratory Standards. (1988).
Hospital Epidemiology 21, 223–6.
Performance Standards for Antimicrobial Susceptibility Testing; 11. Kak, V. & Levine, D. P. (1999). Editorial response: community-
Ninth Informational Supplement: Document M100-S9. NCCLS, acquired methicillin Staphylococcus aureus infections—where do we go from here? Clinical Infectious Diseases 29, 801–2.
26. Asensio, A., Guerrero, A., Quereda, C., Lizan, M. & Martinez-
12. Ayliffe, G. A. J. (1996). Recommendations for the Control of
Ferrer, M. (1996). Colonization and infection with methicillin-resist- Methicillin-Resistant Staphylococcus aureus (MRSA). Division of ant Staphylococcus aureus: associated factors and eradication.
Emerging and other Communicable Diseases Surveillance and Infection Control and Hospital Epidemiology 17, 20–8.
Control, World Health Organisation, Geneva, Switzerland.
27. Boyce, J. M., Landry, M., Deetz, T. R. & Dupont, H. L. (1981).
13. Herwaldt, L. A. (1999). Control of methicillin-resistant Staphylo-
Epidemiologic studies of an outbreak of nosocomial methicillin- coccus aureus in the hospital setting. American Journal of Medicine resistant Staphylococcus aureus infections. Infection Control 2,
14. Monnet, D. (1998). Methicillin-resistant Staphylococcus aureus
28. Ayliffe, G. A. J. (1997). The progressive intercontinental spread
and its relationship to antimicrobial use: possible implications for of methicillin-resistant Staphylococcus aureus. Clinical Infectious control. Infection Control and Hospital Epidemiology 19, 552–9.
Diseases 24, S74–9.
15. Rice, L. (1999). Editorial response: a silver bullet for coloniza-
29. Barie, P. S. (1998). Antibiotic resistant gram positive cocci:
tion and infection with methicillin-resistant Staphylococcus aureus implications for surgical practice. World Journal of Surgery 22, 118–
still eludes us. Clinical Infectious Diseases 28, 1067–70.
16. Schentag, J. J., Hyatt, J. M., Carr, J. R., Paladino, J. A.,
Birmingham, M. C., Zimmer, G. S. et al. (1998). Genesis of methi-
30. Mueller, B. A., Brierton, D. G., Abel, S. R. & Bowman, L. (1994).
cillin-resistant Staphylococcus aureus (MRSA). How treatment of Effect of enteral feeding with ensure on oral bioavailabilities of MRSA infections has selected for vancomycin resistant Entero- ofloxacin and ciprofloxacin. Antimicrobial Agents and Chemo- coccus faecium and the importance of antibiotic management and therapy 38, 2101–5.
infection control. Clinical Infectious Diseases 26, 1204–14.
31. Wright, D. H., Pietz, S. L., Konstantinides, F. N. & Rotschafer,
17. Monnet, D. & Frimodt-Moller, N. (2001). Antimicrobial-drug use
J. C. (2000). Decreased in vitro fluoroquinolone concentrations after and methicillin-resistant Staphylococcus aureus. Emerging Infect- admixture with an enteral feeding formulation. Journal of Parenteral ious Diseases 7, 1–3.
and Enteral Nutrition 24, 42–8.
18. Hill, D. A., Herford, T. & Parratt, D. (1998). Antibiotic usage and
32. Healy, D. P., Brodbeck, M. C. & Clendening, C. E. (1996).
methicillin-resistant Staphylococcus aureus and analysis of causal- Ciprofloxacin absorption is impaired in patients given enteral feed- ity. Journal of Antimicrobial Chemotherapy 42, 676–7.
ings orally and via gastrostomy and jejunostomy tubes. Anti-
microbial Agents and Chemotherapy 40, 6–10.
19. Crowcroft, N. S., Ronveaux, O, Monnet, D. L. & Mertens, R.
(1999). Methicillin-resistant Staphylococcus aureus and anti-
33. Mimoz, O., Binter, V., Jacolot, A., Edouard, A., Tod, M.,
microbial use in Belgian hospitals. Infection Control and Hospital Petitjean, O. et al. (1998). Pharmacokinetics and absolute bioavail- Epidemiology 20, 31–6.
ability of ciprofloxacin administered through a nasogastric tube with Risk factors associated with nosocomical MRSA
continuous enteral feeding to critically ill patients. Intensive Care ceftazidime-resistant Klebsiella pneumoniae following changes in a Medicine 24, 1047–51.
hospital antibiotic formulary. Clinical Infectious Diseases 28, 1062–6.
34. Vincent, J., Teng, R., Pelletier, S. M., Willavize, S. A. &
42. Kaatz, G. W. & Seo, S. M. (1997). Mechanisms of fluoro-
Friedman, H. L. (1998). The bioavailability of nasogastric versus quinolone resistance in genetically related strains of Staphylococcus tablet form oral trovofloxacin in healthy subjects. American Journal aureus. Antimicrobial Agents and Chemotherapy 41, 2733–7.
of Surgery 176, 23S–6S.
43. Ng, E. Y., Trucksis, M. & Hooper, D. C. (1994). Quinolone
35. Fagon, J. Y., Maillet, J. M. & Novara, A. (1998). Hospital
resistance mediated by nor A: physiologic characterization and rela- acquired pneumonia: methicillin resistance and intensive care unit tionship to flqB, a quinolone resistance locus on the Staphylococcus admission. American Journal of Medicine 104, 17S–23S.
aureus chromosome. Antimicrobial Agents and Chemotherapy 38,
36. Pujol, M., Pena, C., Pallares, R., Ayats, J., Ariza, J. & Gudio, F.
(1994). Risk factors for nosocomial bacteremia due to methicillinresistant Staphylococcus aureus. European Journal of Clinical 44. Isaacs, R. D., Kunke, P. J., Cohen, R. L. & Smith, J. W. (1988).
Microbiology and Infectious Diseases 13, 96–102.
Ciprofloxacin resistance in epidemic methicillin-resistant Staphylo-
coccus aureus. Lancet ii, 843.
37. Witte, W., Braulke, C., Heuck, D. & Cuny, C. (1994). Analysis
of nosocomial outbreaks with multiply and methicillin-resistant
45. Manhold, C., von Rolbicki, U., Brase, R., Timm, J., von Pritz-
Staphylococcus aureus (MRSA) in Germany: implications for buer, E., Heimesaat, M. et al. (1998). Outbreaks of Staphylococcus hospital hygiene. Infection 22, S128–34.
aureus infections during treatment of late onset pneumonia with 38. Rosdahl, V. T. & Knudsen, A. M. (1991). The decline of
ciprofloxacin in a prospective, randomized study. Intensive Care methicillin resistance among Danish Staphylococcus aureus Medicine 24, 1327–30.
strains. Infection Control and Hospital Epidemiology 12, 83–8.
46. Gottlieb, T. & Mitchell, D. (1998). The independent evolution of
39. Law, M. R. & Gill, O. N. (1988). Hospital-acquired infection with
resistance to ciprofloxacin, rifampin, and fusidic acid in methicillin- methicillin-resistant and methicillin-sensitive staphylococci. Epidemi- resistant Staphylococcus aureus in Australian teaching hospitals ology and Infection 101, 623–9.
(1990–1995). Journal of Antimicrobial Chemotherapy 42, 67–73.
40. Acar, J. F. & Goldstein, F. W. (1997). Trends in bacterial resist-
47. Venezia, R. A., Damaracki, B. E., Evans, A. M., Preston, K. E.
ance to fluoroquinolones. Clinical Infectious Diseases 24, S67–73.
& Graffunder, E. M. (2001). Selection of high-level oxacillin resist- 41. Landman, D., Chockalingam, M. & Quale, J. (1999). Reduction
ance in heteroresistant Staphylococcus aureus by fluoroquinolone in the incidence of methicillin-resistant Staphylococcus aureus and exposure. Journal of Antimicrobial Chemotherapy 48, 375–81.

Source: http://mrsa-net.nl/files/de/file-bron-ant-66-2-Graffunder.pdf