Community‐Acquired Infection With Healthcare‐Associated Methicillin‐Resistant Staphylococcus aureus: The Role of Home Nursing Care
Objective. To better understand the role of indirect transmission in community‐acquired infection with methicillin‐resistant Staphylococcus aureus (MRSA).
Design. Prospective case‐control study.
Setting. A French teaching hospital.
Patients. A total of 198 case patients and 198 control patients with MRSA or methicillin‐susceptible S. aureus infection diagnosed between April 2002 and July 2003.
Results. Multivariate analysis showed a highly significant independent link between MRSA infection at admission and prior receipt of home nursing care (odds ratio [OR], 3.7; 
). Other independent risk factors were prior hospitalization (OR, 3.8; 
), transfer from another institution (OR, 2.3; 
), and age older than 65 years (OR, 1.6; 
). Prior home nursing care showed a frequency dose‐response relationship. Eleven MRSA‐infected patients had had home nursing procedures but no hospital stay in the previous 3 years. These patients’ MRSA strains were related to the prevalent MRSA clone currently spreading in French hospitals.
Conclusion. Home nursing care appears to be an independent risk factor for MRSA acquisition in the community. The reservoir probably consists of MRSA carriers discharged from the hospital. Community nurses seem to be a potential vector.
Received November 20, 2005; accepted February 23, 2006; electronically published October 20, 2006.
Methicillin‐resistant Staphylococcus aureus (MRSA) is responsible for 2%‐40% of all hospital‐acquired staphylococcal infections in Europe, with a marked North‐South gradient of increasing frequency.1 Hospital‐acquired MRSA infection is associated with country‐ or continent‐specific MRSA clones.2,3 This clonality reflects the success of particular genotypes adapted to the hospital environment. MRSA is also emerging as a community‐acquired pathogen. The MRSA strains that are spreading in the community are clonally different from MRSA strains transmitted within hospitals, and they harbor specific virulence factors, such as the Panton‐Valentine leukocidin genes.4,5 Healthcare‐associated MRSA (HA‐MRSA) strains can also be carried into the community by people without symptoms of infection and can be spread by person‐to‐person contact. Even when such strains are detected in the community, they are often referred to as HA‐MRSA.
HA‐MRSA strains tend to infect older patients (median age, 68 years6) who usually have underlying diseases (eg, diabetes mellitus, cancer, and chronic skin disease) and/or indwelling urinary tract or intravascular devices in place.7‐11 MRSA infections seen in the community can be acquired either directly in hospitals or long‐stay institutions or indirectly by contact with an MRSA carrier, such as a family member working in a hospital, a family member with a previous stay in the hospital, a general practitioner, or a community nurse.12‐16 Reports of indirect transmission of MRSA strains are anecdotal, whereas large series of cases of direct transmission have been reported.17,18 To better understand the role of indirect transmission in community‐acquired MRSA infection, we performed a prospective case‐control study to determine all of the potential factors of MRSA transmission in the community.
Methods
Study Population
This prospective case‐control study took place in Amiens teaching hospital, a 1,650‐bed facility in Amiens, France, between April 2002 and July 2003. Adult patients older than 18 years were eligible if they had a diagnostic sample obtained within 48 hours of admission that tested positive for S. aureus. A case‐control study was conducted. Case patients were defined as patients infected with MRSA, and control patients were defined as patients infected with methicillin‐susceptible S. aureus. Diagnoses of infections were made by hospital physicians who examined patients within the first 48 hours after admission. Diagnosis of infections was defined by means of the MeSH terms “soft tissue infection” or “abscess,” “bacterial pneumonia” or “chronic bronchitis,” “urinary tract infection,” “bacteremia,” “central nervous system bacterial infection,” and “infectious arthritis” (the National Center for Biotechnology Information MeSH database is available at: http://www.ncbi.nlm.nih.gov/). Controls were enrolled with cases in a ratio of 1:1, with less than 3 months separating the period of hospitalization for each member of the pair. Patients were ineligible for the study if they were younger than 18 years, if they were receiving hemodialysis for chronic renal failure, and if S. aureus was detected by nasal or axillary screening in the absence of symptoms of infection.
Conduct of the Study
After giving their informed consent to participate in the study, patients completed a questionnaire issued by a specially trained assistant. Depending on their health status on admission, patients were contacted either during the hospital stay or after returning home. The interview, which lasted up to 15 minutes, took place either face to face or over the telephone. If patients had dementia or memory problems, we questioned the closest relatives. If a patient died before completing the questionnaire, we questioned their family practitioner. The diagnostic samples were acquired because of suspected bacteremia, urogenital infection, skin and soft‐tissue infection, respiratory tract infection, arthritis, and central nervous system infection.
Analytical Variables
The dependent variable was the S. aureus isolate's methicillin resistance status. The independent variables were the study participants' sociodemographic characteristics, medical or surgical history, reasons for admission, findings from an exhaustive 3‐year institutional history, previous receipt of medical or paramedical care in the community, and possible contacts with individuals at high risk for MRSA carriage (ie, close family members previously infected by MRSA and family members working in the hospital setting). The principal variables were those that assessed different healthcare services attendance. If a patient had previously received treatment in a hospital, private clinic, or nursing home, we sought to establish the period and duration of their hospitalization and the ward concerned. If community health care had been administered during the 3 years that preceded admission, we also sought to determine the number, duration, and period of interventions and the professional(s) involved (ie, physician, specialist, nurse, and/or physiotherapist). Nursing home care was categorized as parenteral infusion or injections, wound dressing, and urinary catheter changes. If family contacts at risk were found, we recorded the family relationship, the frequency of contacts, and the duration of contact before the patient was admitted.
Microbiological Procedures
Most samples were of urine, sputum, bronchoalveolar lavage fluid, abscess fluid, swabs of a tissue infection site, blood, synovial fluid, or cerebrospinal fluid. S. aureus was identified by a test (bioMérieux) used to detect its ability to produce coagulase. Susceptibility to methicillin was determined by the standard disk method on Mueller‐Hinton agar that contained 2% sodium chloride (bioMérieux). Growth inhibition was interpreted according to the standard recommendations of the Comité de l’Antibiogramme de la Société Française de Microbiologie.19 Pulsed‐field gel electrophoresis (PFGE) was performed after restriction endonuclease digestion of chromosomal DNA with SmaI, as recommended by the manufacturer (Bio‐Rad France).20
Bacteriological and genomic analysis was performed for S. aureus isolates recovered from a subgroup of 20 patients, including 13 randomized patients with different durations of past hospitalization, 6 randomized patients who had never been in a hospital or other official healthcare centers, and 1 control infected with methicillin‐susceptible S. aureus. Multilocus sequence typing was performed as described by Enright et al.21 for determination of sequence types (STs). The ST was assigned according to the multilocus ST database (available at: http://www.mlst.net). The accessory gene regulator (agr) allele group (alleles 1‐4) was determined by multiplex polymerase chain reaction.22 The type of staphylococcal chromosomal cassette (SCC) encoding resistance to methicillin (SCCmec) was determined according to the method of Oliveira and de Lencastre.23
Statistical Analysis
The questionnaire results were recorded with Epi Info software, version 6.04 fr (Centers for Disease Control and Prevention), and statistical analyses was performed with SAS software, version 8.0 (SAS Institute). A descriptive analysis was used to study the distribution of continuous variables (ie, means, medians, SDs, ranges, and quartiles) and the frequencies of categorical variables. We then performed a case‐control study. Means were compared by using Student's t test or, when appropriate, the Wilcoxon nonparametric test. Percentages were compared with the Pearson's χ2 test or, when appropriate, Fisher's exact test. A P value of less than .05 was considered to denote statistical significance.
Independent variables with a P value of .20 or less after univariate analysis logistic regression were included in a multiple logistic regression model based on a stepwise ascending method. The fit of the model was studied with the likelihood ratio test. The stability of the final model was then tested by sensitivity analyses and by the Hosmer‐Lemeshow goodness‐of‐fit test.24 A P value of less than .05 was considered to denote statistical significance.
Results
Three hundred ninety‐six infected patients (198 cases and 198 controls) were included in the study. We failed to contact 18 cases and 17 controls, and 5 cases and 8 controls refused to fill out the questionnaire. Hence, 175 cases (88.4%) and 173 controls (87.4%) were interviewed. A total of 139 patients (39.9%) were enrolled from medicine wards, 114 (32.8%) were from surgery wards, 50 (14.4%) were from intensive care units, 30 (8.6%) were from geriatric‐rehabilitation departments, and 15 (4.3%) were from gynecology wards. Demographic data and results of univariate analysis of the types of infections and the risk factors for the 2 groups are listed in Table 1.
Univariate analysis (Table 1) showed that the cases were older than the controls (mean age [±SD], 
and 57 ± 20 years; 
). Cases were more frequently admitted to geriatric‐rehabilitation centers (odds ratio [OR], 3.0; P = .008) and less frequently admitted to intensive care units (OR, 0.4; 
). As expected, risk factors classically associated with infection due to HA‐MRSA strains were significantly more common among the cases, including medical antecedents (OR, 1.9; P = .01), surgical antecedents (OR, 1.8; P = .01), prosthesis use (OR, 2.4; 
), urinary tract infection (OR, 3.3; 
), stay at a long‐term care facility (OR, 2.5; 
), and transfer from another healthcare institution (OR, 2.5; 
) (Table 1). No differences were found for follow‐up observation at a community healthcare facility or for performance of outpatient surgery. MRSA infection tended to be associated with recent receipt of antimicrobial therapy (OR, 1.5; 
). No differences or trends were found for a specific family of antibiotics (eg, quinolones).
MRSA infection was significantly associated with prior hospitalization (OR, 2.7; 
) and receipt of home nursing care within the preceding 3 years (OR, 2.6; 
). The cumulative number of contacts with community healthcare nurses was nearly twice as high among the cases as among the controls (
). A dose‐response relationship was also observed, confirming the association between MRSA infection and receipt of home nursing care (Table 1): the higher the number of home nursing procedures, the higher the risk of MRSA infection. We identified a group of 11 cases for whom home nursing procedures had been performed but who had not been admitted to the hospital during the 3 years preceding this study (and therefore had not been transferred from another institution).
A positive interaction effect was sought to determine links between the different risk factors. Positive interaction effects were identified between receipt of home nursing care and medical (
) and surgical (
) antecedents, prior hospitalization in surgical wards (
), and age older than 65 years.
Multivariate analysis (Table 2) showed a strong and highly significant independent association between MRSA infection and prior receipt of home nursing care (OR, 3.7; 
). This was a risk factor independent of prior hospitalization, transfer from another institution, and age older than 65 years. There was also a positive interaction effect between prior receipt of home nursing care and prior hospitalization in a surgical ward.
We then compared the bacteriological characteristics of 19 randomly selected MRSA isolates from cases with a history of hospitalization (13 cases), history of hospitalization and receipt of home nursing care (15 cases), and contact with a healthcare worker (1 case). We considered isolates whose PFGE pattern was more than 80% similar as belonging to the same PFGE type. According to this definition, the 19 isolates belonged to 2 PFGE types (A and B) (Figure). Isolates of either PFGE type were considered to belong to the same clone when they had the same ST (ST8), the same agr allele (allele 1), and a SCCmec type IVA element. This clone corresponds to the main clone of the hospital‐acquired MRSA strains spreading in France (J. Etienne, unpublished data).
Figure. Dendrogram comparing bacteriological characteristics of methicillin‐resistant Staphylococcus aureus (MRSA) isolates, including sequence type (ST), agr type, and presence of the staphylococcal chromosomal cassette encoding resistance to methicillin (SCCmec) type, recovered from case patients with a history of hospitalization (HPH), HPH and home nursing care (HNC), or healthcare worker contact (HCW). Results in bold correspond to patients with HNC only. The methicillin‐susceptible S. aureus (MSSA) strain was recovered from a control patient. ND, not determined; NT, not typable; PFGE, pulsed‐field gel electrophoresis; +, present; −, absent.
Discussion
This prospective case‐control study of 348 patients identifies receipt of home nursing care as an independent risk factor for MRSA infection in the community. This association persisted after adjustment for classic risk factors (OR, 3.7) and was highly statistically significant (
). Moreover, there was a clear dose‐response relationship between the risk of MRSA acquisition and the number of contacts with home healthcare nurses. The MRSA strains acquired in the community resembled the prevalent clone of HA‐MRSA circulating in France. Thus, it appeared that MRSA‐infected patients were infected by strains of hospital origin, whatever the mode of acquisition. Community nurses appeared to be potential vectors of HA‐MRSA strains among patients with no prior hospitalization.
We confirmed recognized risk factors for HA‐MRSA acquisition, such as prior hospitalization and receipt of long‐term institutional care,7‐11 but not indirect contacts with healthcare workers (eg, a family member working in a hospital).17‐18 The study population investigated herein was large compared with populations described in other published studies of risk factors for HA‐MRSA acquisition.11,15,25 The prospective design of the study meant that there were few missing data, and recall bias was minimized by pooling information collected from the patients themselves, their relatives, and their general practitioners. The final model with 348 patients was stable; that is, this population was poorly sensitive to the addition or withdrawal of observations (
, by the residual χ2 test; 
by the Hosmer‐Lemeshow test).
Several potential biases in the design of our study should be discussed. Our patients with community‐acquired MRSA infection acquired in the community represent only those who were admitted to the hospital and who were only part of the general population infected with MRSA. Hence, our results demonstrate the role of home nursing care in the transmission of HA‐MRSA in a group of patients for whom the risk of MRSA infection during hospitalization is high and cannot be strictly extrapolated to the general population. However, the results allowed us to find significant differences, meaningful both epidemiologically and biologically.
The association between receipt of home nursing care and MRSA infection has previously been reported in several studies.12,25 In a study by Troillet et al.12 performed in 1998, a total of 387 patients were screened within 24 hours after admission. Ten patients were found to be colonized or infected with MRSA, and 5 of the 6 patients for whom the relevant information was available had received home health care within the past year. In our study, logistic regression analysis identified performance of home nursing procedures as an independent risk factor for MRSA infection. In the French system, nurses provide care at the patient’s home and work independently of healthcare centers.
Transient MRSA carriage on the hands of healthcare workers is the most important route of MRSA dissemination from person to person.26 The mean duration of MRSA carriage is approximately 6 months, but this can increase to 40 months in some patients, such as those with diabetes mellitus or chronic wounds.27 Patients who acquire MRSA during hospitalization are not necessarily identified at the time of discharge, and if they are, this information is rarely provided to community nurses who care for these patients. Community healthcare workers are possibly less aware of policies of nosocomial infection control. The high rate of MRSA endemicity among French hospitals (approximately 30%–40% of all French hospitals),28 together with the rapid development of ambulatory health care, creates the conditions required for HA‐MRSA spread to the community. A similar pattern was seen in the 1940s and 1950s with penicillin resistance: penicillin resistance among S. aureus strains became frequent among isolates recovered from hospitalized patients before these strains were found to be spreading to the community.29 It is noteworthy that all of the HA‐MRSA strains characterized in this study belonged to the main French hospital MRSA clone (agr1, ST8, and SCCmec‐IVA). This study suggests that infection control procedures, particularly hand hygiene based on alcohol rubbing, should be reinforced among community healthcare workers. It might also be helpful if these personnel were informed of the MRSA carrier status of their patients after the patients have been discharged from the hospital.
Acknowledgments
We thank Anna Schuh and Alexandra Ford, for reviewing the manuscript; David Young, for editing the manuscript; Pierre Duhaut, for his statistical expertise; and Sonia Beltran, Domitille Beuvin, Anne‐Sophie Bonnard, Nathalie Decaux, Hyacinthe Duwat, Marie Hélène Favre, Celia Hennick, Katy‐Anne Walczak, Michèle Bes, Hélène Meugnier, and Cyrille Pila, for their logistical assistance.
References
- 1. Tiemersma EW, Bronzwaer SL, Lyytikainen O, et al. Methicillin‐resistant Staphylococcus aureus in Europe, 1999‐2002. Emerg Infect Dis 2004; 10:1627‐1634.
- 2. Heym B, Le Moal M, Armand‐Lefevre L, Nicolas‐Chanoine MH. Multilocus sequence typing (MLST) shows that the ’Iberian’ clone of methicillin‐resistant Staphylococcus aureus has spread to France and acquired reduced susceptibility to teicoplanin. J Antimicrob Chemother 2002; 50:323‐329.
- 3. Oliveira DC, Tomasz A, de Lencastre H. Secrets of success of a human pathogen: molecular evolution of pandemic clones of methicillin‐resistant Staphylococcus aureus. Lancet Infect Dis 2002; 2:180‐189.
- 4. Vandenesch F, Naimi T, Enright MC, et al. Community‐acquired methicillin‐resistant Staphylococcus aureus carrying Panton‐Valentine leukocidin genes: worldwide emergence. Emerg Infect Dis 2003; 9:978‐984.
- 5. Francis JS, Doherty MC, Lopatin U, et al. Severe community‐onset pneumonia in healthy adults caused by methicillin‐resistant Staphylococcus aureus carrying the Panton‐Valentine leukocidin genes. Clin Infect Dis 2005; 40:100‐107.
- 6. Naimi TS, LeDell KH, Como‐Sabetti K, et al. Comparison of community‐ and health care‐associated methicillin‐resistant Staphylococcus aureus infection. JAMA 2003; 290:2976‐2984.
- 7. Salgado CD, Farr BM, Calfee DP. Community‐acquired methicillin‐resistant Staphylococcus aureus: a meta‐analysis of prevalence and risk factors. Clin Infect Dis 2003; 36:131‐139.
- 8. Charlebois ED, Bangsberg DR, Moss NJ, et al. Population‐based community prevalence of methicillin‐resistant Staphylococcus aureus in the urban poor of San Francisco. Clin Infect Dis 2002; 34:425‐433.
- 9. Abudu L, Blair I, Fraise A, Cheng KK. Methicillin‐resistant Staphylococcus aureus (MRSA): a community‐based prevalence survey. Epidemiol Infect 2001; 126:351‐356.
- 10. Hussain FM, Boyle‐Vavra S, Bethel CD, Daum RS. Current trends in community‐acquired methicillin‐resistant Staphylococcus aureus at a tertiary care pediatric facility. Pediatr Infect Dis J 2000; 19:1163‐1166.
- 11. Rezende NA, Blumberg HM, Metzger BS, Larsen NM, Ray SM, McGowan JE Jr. Risk factors for methicillin‐resistance among patients with Staphylococcus aureus bacteremia at the time of hospital admission. Am J Med Sci 2002; 323:117‐123.
- 12. Troillet N, Carmeli Y, Samore MH, et al. Carriage of methicillin‐resistant Staphylococcus aureus at hospital admission. Infect Control Hosp Epidemiol 1998; 19:181‐185.
- 13. Shopsin B, Mathema B, Martinez J, et al. Prevalence of methicillin‐resistant and methicillin‐susceptible Staphylococcus aureus in the community. J Infect Dis 2000; 182:359‐362.
- 14. Calfee DP, Durbin LJ, Germanson TP, Toney DM, Smith EB, Farr BM. Spread of methicillin‐resistant Staphylococcus aureus (MRSA) among household contacts of individuals with nosocomially acquired MRSA. Infect Control Hosp Epidemiol 2003; 24:422‐426.
- 15. Lesens O, Hansmann Y, Brannigan E, et al. Healthcare‐associated Staphylococcus aureus bacteremia and the risk for methicillin resistance: is the Centers for Disease Control and Prevention definition for community‐acquired bacteremia still appropriate? Infect Control Hosp Epidemiol 2005; 26:204‐209.
- 16. Friedman ND, Kaye KS, Stout JE, et al. Health care–associated bloodstream infections in adults: a reason to change the accepted definition of community‐acquired infections. Ann Intern Med 2002; 137:791‐797.
- 17. Eveillard M, Martin Y, Hidri N, Boussougant Y, Joly‐Guillou ML. Carriage of methicillin‐resistant Staphylococcus aureus among hospital employees: prevalence, duration, and transmission to households. Infect Control Hosp Epidemiol 2004; 25:114‐120.
- 18. Wagenvoort JH, De Brauwer EI, Sijstermans ML, Toenbreker HM. Risk of re‐introduction of methicillin‐resistant Staphylococcus aureus into the hospital by intrafamilial spread from and to healthcare workers. J Hosp Infect 2005; 59:67‐68.
- 19. Comité de l’Antibiogramme de la Société Française de Microbiologie. Zone sizes and IMC break‐points and non‐fastidious organisms. Clin Microbiol Infect 1996; 2(Suppl 1):S1‐S49.
- 20. Goering RV, Winters MA. Rapid method for epidemiological evaluation of gram‐positive cocci by field inversion gel electrophoresis. J Clin Microbiol 1992; 30:577‐580.
- 21. Enright MC, Day NP, Davies CE, Peacock SJ, Spratt BG. Multilocus sequence typing for characterization of methicillin‐resistant and methicillin‐susceptible clones of Staphylococcus aureus. J Clin Microbiol 2000; 38:1008‐1015.
- 22. Jarraud S, Mougel C, Thioulouse J, et al. Relationships between Staphylococcus aureus genetic background, virulence factors, agr groups (alleles), and human disease. Infect Immun 2002; 70:631‐641.
- 23. Oliveira DC, de Lencastre H. Multiplex PCR strategy for rapid identification of structural types and variants of the mec element in methicillin‐resistant Staphylococcus aureus. Antimicrob Agents Chemother 2002; 46:2155‐2161.
- 24. Lemeshow S, Hosmer DW Jr. A review of goodness of fit statistics for use in the development of logistic regression models. Am J Epidemiol 1982; 115:92‐106.
- 25. Tambyah PA, Habib AG, Ng TM, Goh H, Kumarasinghe G. Community‐acquired methicillin‐resistant Staphylococcus aureus infection in Singapore is usually “healthcare associated.” Infect Control Hosp Epidemiol 2003; 24:436‐438.
- 26. Pittet D, Dharan S, Touveneau S, Sauvan V, Perneger TV. Bacterial contamination of the hands of hospital staff during routine patient care. Arch Intern Med 1999; 159:821‐826.
- 27. MacKinnon MM, Allen KD. Long‐term MRSA carriage in hospital patients. J Hosp Infect 2000; 46:216‐221.
- 28. Stefani S, Varaldo PE. Epidemiology of methicillin‐resistant staphylococci in Europe. Clin Microbiol Infect 2003; 9:1179‐1186.
- 29. Chambers HF. The changing epidemiology of Staphylococcus aureus? Emerg Infect Dis 2001; 7:178‐182.