Routine Screening for Methicillin‐Resistant Staphylococcus aureus Among Patients NewlyAdmitted to an Acute Rehabilitation Unit  • Author(s): Farrin A. Manian , MD, MPH; Diane Senkel , RN, CIC; Jeanne Zack , RN, BSN, CIC;Lynn Meyer , RN, MPH, CICSource: Infection Control and Hospital Epidemiology, Vol. 23, No. 9 (September 2002), pp. 516-519Published by: The University of Chicago Press on behalf of The Society for Healthcare Epidemiologyof AmericaStable URL: http://www.jstor.org/stable/10.1086/502099 .

Accessed: 14/05/2014 23:47

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

The University of Chicago Press and The Society for Healthcare Epidemiology of America are collaboratingwith JSTOR to digitize, preserve and extend access to Infection Control and Hospital Epidemiology.

http://www.jstor.org

This content downloaded from 91.229.248.208 on Wed, 14 May 2014 23:47:24 PMAll use subject to JSTOR Terms and Conditions

516 INFECTION CONTROL AND HOSPITAL EPIDEMIOLOGY September 2002

ROUTINE SCREENING FOR METHICILLIN-RESISTANT

STAPHYLOCOCCUS AUREUS AMONG PATIENTS NEWLY

ADMITTED TO AN ACUTE REHABILITATION UNIT

Farrin A. Manian, MD, MPH; Diane Senkel, RN, CIC; Jeanne Zack, RN, BSN, CIC; Lynn Meyer, RN, MPH, CIC

BACKGROUND: Following an outbreak of methicillin-resistant Staphylococcus aureus (MRSA) infection in our acuterehabilitation unit in 1987, all patients except in-house transfers(because of their low prevalence of MRSA colonization) under-went MRSA screening cultures on admission.

OBJECTIVES: To better characterize the current profile ofpatients with positive MRSA screening cultures at the time ofadmission to our acute rehabilitation unit, and to determine the rel-ative yield of nares, perianal, and wound screening cultures in thispopulation.

METHODS: Prospective chart review with ongoing activesurveillance for infections associated with the acute rehabilitationunit.

RESULTS: The rate of MRSA isolation from one or more

body sites increased significantly from 5% (1987–1988) to 12%(1999–2000) (P = .0009) for newly admitted patients and from 0% to7% (P < .0001) for in-house transfers. A negative nares culture washighly predictive (98%) of a negative perianal culture. Prior historyof MRSA infection or colonization and transfer from outsidesources were independently associated with positive MRSAscreening cultures.

CONCLUSION: The rate of MRSA isolation from screen-ing cultures of newly admitted patients, including in-house trans-fers, has increased significantly during the past decade in ouracute rehabilitation unit. When paired with nares cultures, perianalcultures were of limited value in this patient population (InfectControl Hosp Epidemiol 2002;23:516-519).

Methicillin-resistant Staphylococcus aureus (MRSA)is a frequent cause of infection among patients admitted toacute rehabilitation units.1 Although the optimal approachto the control of MRSA in institutions has yet to be defined,screening cultures for early identification of carriers, withpossible eradication of the carriage state, have been sug-gested.2,3 Following an outbreak of MRSA infection on theacute rehabilitation unit of our medical center in 1987, a pol-icy of routine MRSA screening cultures of the nares, theperianal area, and open cutaneous sites (including gastros-tomy and tracheostomy sites) of patients admitted fromoutside institutions was implemented; in-house transferswere exempted because of the low (0%) rate of positivescreening cultures among this patient population at thattime.4 Published studies addressing the problem of MRSAin acute rehabilitation units have either reported data pri-marily from patients with spinal cord injury5,6 or failed toroutinely perform surveillance cultures for MRSA at thetime of patient admission.1 Because of a perceived increasein the rate of nosocomial MRSA infection among our hos-pitalized patients during the past decade, we reevaluatedthe policy of restricting MRSA screening to patients newlyadmitted from sources other than our affiliated hospital.Furthermore, we studied patient characteristics associatedwith positive MRSA screening cultures and the relative

yield of MRSA cultures from various body sites and com-pared the prevalences of MRSA colonization amongpatients newly admitted to our acute rehabilitation unit dur-ing the past decade.

METHODS

St. John’s Mercy Medical Center is an 859-bed, com-munity teaching hospital located in Creve Coeur, a suburbof St. Louis, Missouri. Located within the medical centercomplex is a 45-bed acute rehabilitation unit with 14 pri-vate rooms and an average admission rate of 20 patientsper month. From July 1, 1999, through May 20, 2000, allnewly admitted patients underwent MRSA screening cul-tures of the nares, the perianal area, and any existingwounds, including gastrostomy and tracheostomy sites.Patients with one or more positive MRSA cultures wereroutinely placed in a private room or cohorted with otherpatients with a positive MRSA culture from any body site,and received mupirocin ointment to the colonized site orsites twice daily for 1 week in an attempt to eradicateMRSA carriage. Cultures were obtained using sterileswabs (Culturette, Baxter Healthcare Corp., Deerfield,IL). Mannitol salt agar supplemented with oxacillin (4µg/mL) was used for screening of MRSA growth. S.aureus identification was performed with the latex aggluti-

The authors are from the Department of Infection Control, St. John’s Mercy Medical Center, St. Louis, Missouri.Address reprint requests to Farrin A. Manian, MD, MPH, 621 S. New Ballas, St. Louis, MO 63141.The authors thank Marlene Northway and Jennifer Manian for their assistance with and diligence in data entry.

ABSTRACT

This content downloaded from 91.229.248.208 on Wed, 14 May 2014 23:47:24 PMAll use subject to JSTOR Terms and Conditions

Vol. 23 No. 9 MRSA IN AN ACUTE REHABILITATION UNIT 517

nation test (Staphaurex, Murex Biotech, Ltd., Dartford,United Kingdom).

Data regarding patient characteristics and cultureresults were obtained by prospective chart review. Becausethis project was part of the ongoing quality improvementactivities of our medical center with potential benefit topatients, no formal written consent was required; however,patients had the right to refuse screening cultures.Ongoing monitoring for newly acquired infections in theacute rehabilitation unit was performed by an infection con-trol professional. “Acute rehabilitation unit–associated”infection and colonization were defined on the basis of theCenters for Disease Control and Prevention criteria fornosocomial infection.7 “Prior antibiotic treatment” wasdefined as documentation of one or more doses of systemicantibiotics up to 2 weeks prior to patient admission to theacute rehabilitation unit. “Colonization” was defined as theisolation of MRSA from one or more body sites withoutclinical evidence of active infection. “History of MRSA” wasdefined as documentation of MRSA infection or coloniza-tion up to 1 year prior to admission.

Statistical analysis was performed using InStat(GraphPad Software Inc., San Diego, CA) and Statistica(StatSoft, Inc., Tulsa, OK) software packages. Chi-square(with Yates correction for 2 � 2 tables) and Fisher’s exacttests were used for comparison of categoric data. TheStudent’s t test was used for comparison of continuous data.Multiple logistic regression analysis was employed fordetermination of factors independently associated withMRSA isolation from one or more screened body sites,including in the model those variables associated withMRSA isolation at a P value of less than .2 in the univariateanalysis. A P value of less than .05 was considered statisti-cally significant.

RESULTS

From July 1, 1999, through May 20, 2000, there were427 admissions to the acute rehabilitation unit, represent-ing 419 unique patients with a mean age of 53 years(range, 15 to 91 years); 250 (59%) were male. Mean dura-tion of stay in the acute rehabilitation unit was 21 days(range, 1 to 209 days). The most common primary diagno-sis at the time of admission was brain injury (traumatic orassociated with cerebrovascular accident, 204; 48%), fol-lowed by multiple fractures (51; 12%), “ambulation dys-function” (42; 10%), spinal cord injury (33; 8%), post-burninjuries (30; 7%), and post-orthopedic surgery (28; 7%).The remaining diagnoses (8%) fell into the miscellaneouscategory, including altered mental status and multiple scle-rosis. The referral sources for patients were St. John’sMercy Medical Center (307; 72%), acute care hospitals (81;19%), home (20; 5%), and skilled nursing facilities (19; 4%).History of prior MRSA infection or colonization was docu-mented in 27 (6.3%) patients and prior antibiotic treatmentwas documented in 180 (42%) patients.

Of 427 consecutive patients admitted, 411 (96.3%)underwent screening cultures of the nares, the perianalarea, and wound or wounds (when present); 51 (12%) had

MRSA from cultures of one or more sites. The frequency ofpositive MRSA screening cultures by body site is listed inTable 1. The rate of MRSA isolation from gastrostomy ortracheostomy sites was significantly higher than that fromnares. Of 11 positive gastrostomy site and 9 positive tra-cheostomy site cultures, 3 (27%) and 0 (0%) sites, respec-tively, were considered infected by the admitting physician.Of 6 patients with positive miscellaneous wound cultures(excluding gastrostomy and tracheostomy sites), 2 (33%)were considered infected.

Of 31 patients with a positive MRSA nares culture, 13(42%) also had a positive MRSA perianal culture, comparedwith 9 (2%) of 380 patients with a negative nares culture buta positive perianal culture (P < .0001). The positive and neg-ative predictive values of MRSA nares culture for perianalculture were 42% (95% confidence interval [CI95], 24% to60%) and 98% (CI95, 96% to 99%), respectively.

Of 16 patients with a positive MRSA nares cultureand a concurrent wound or wounds (including gastrosto-my and tracheostomy sites), 10 (67%) had a positiveMRSA wound culture, compared with 16 (16.7%) of 96patients with a negative nares culture and a concurrentlycolonized or infected wound or wounds (P = .0001). Thepositive and negative predictive values of MRSA naresculture for wound culture were 67% (CI95, 38% to 88%) and83% (CI95, 74% to 90%), respectively.

The rate of positive MRSA cultures by site and select-ed patient attributes is listed in Table 2. A positive MRSAculture from one or more body sites was significantly asso-ciated with a prior history of MRSA infection or coloniza-tion, prior antibiotic therapy, and transfer from sources outside of St. John’s Mercy Medical Center (particularlyoutside hospitals and skilled nursing facilities); age, gen-der, and admitting diagnosis were not independently asso-ciated with positive screening cultures (data not shown).

Multiple logistic regression analysis of factors signif-icantly associated with a positive MRSA culture from anysite showed only history of prior MRSA colonization or

TABLE 1FREQUENCY OF POSITIVE METHICILLIN-RESISTANT

STAPHYLOCOCCUS AUREUS SCREENING CULTURES AMONG PATIENTS

NEWLY ADMITTED TO THE ST. JOHN’S MERCY MEDICAL CENTER

REHABILITATION UNIT DURING 1999–2000 BY BODY SITE

No. of Patients No. of MRSA-

Undergoing PositiveBody Site Cultures Cultures (%%)

Nares 411 31 (7.3)*,†

Perianal 411 22 (5.4)Gastrostomy tube site 49 11 (22)*Tracheostomy tube site 38 9 (24)†

Miscellaneous open wounds 57 6 (10.5)

MRSA = methicillin-resistant Staphylococcus aureus.*P = .002; odds ratio, 3.5 (95% confidence interval, 1.7 to 7.6).†P = .003; odds ratio, 3.8 (95% confidence interval, 1.7 to 8.8).

This content downloaded from 91.229.248.208 on Wed, 14 May 2014 23:47:24 PMAll use subject to JSTOR Terms and Conditions

518 INFECTION CONTROL AND HOSPITAL EPIDEMIOLOGY September 2002

infection and source facility as statistically significant inde-pendent predictors of a positive MRSA culture at the timeof admission to the acute rehabilitation unit (P < .00001[odds ratio, 32; CI95, 11 to 95] and P = .002 [odds ratio, 1.2;CI95, 1.03 to 1.4], respectively).

Of 411 patients who underwent MRSA screeningcultures, 4 (1%) subsequently acquired MRSA infectionduring their stay on the acute rehabilitation unit: 1 patientwith positive and 3 patients with negative MRSA screen-ing cultures at the time of admission, representing a 2%and a 0.8% rate of MRSA infection in their respectivegroups (P = .4). The single acute rehabilitation unit–asso-ciated MRSA infection in the patient with a positive MRSAscreening culture occurred 110 days following his admis-sion, whereas the 3 acute rehabilitation unit–associatedMRSA infections among patients with negative screeningcultures developed a mean of 7 days (range, 3 to 12 days)following their admission. Patients with MRSA at one ormore sites at the time of their admission had a signifi-cantly longer stay in the acute rehabilitation unit than didpatients with negative screening cultures (mean, 28.6 vs20.4 days; P = .006).

A comparison of the rates of positive MRSA culturesfrom one or more body sites in the current study with thoseof a previous study performed in the same unit from April1987 through December 19884 revealed an increase in therate of MRSA isolation among newly admitted patients: 51(12%) of 411 compared with 23 (5.4%) of 420, respectively (P< .0009; odds ratio, 2.3; CI95, 1.4 to 3.9). For in-house trans-fers, the rate of positive MRSA cultures from one or morebody sites rose from 0 of 240 to 22 (7%) of 295 (P < .0001;

odds ratio, 39.6; CI95, 2.4 to 657), whereas for patientsadmitted from outside sources this rate increased from 23(13%) of 180 to 29 (25%) of 116 (P = .008; odds ratio, 2.4;CI95, 1.2 to 4.2) during the same periods.

DISCUSSION

Because many patients admitted to acute rehabilita-tion units are from hospital settings, an increase in the rateof MRSA colonization or infection among hospitalizedpatients is likely to be associated with increasing “importa-tion” of MRSA carriers to rehabilitation facilities. Indeed,we found a 200% increase in the prevalence of MRSA isola-tion from patients newly admitted to our acute rehabilita-tion unit between the 1987–1988 and 1999–2000 study peri-ods; this is similar to the 260% increase in MRSA isolationreported from hospitals participating in the InternationalNetworks for the Study and Prevention of EmergingAntimicrobial Resistance from 1990 to 1997.8 Furthermore,we found that the increase in the rate of MRSA isolationfrom designated body sites of newly admitted patients wasnot limited to those transferred from outside facilities; thisrate also increased significantly among patients transferredwithin our own hospital (from 0% to 7%), suggesting that wemay no longer be able to automatically exclude suchpatients from routine MRSA screening in our acute reha-bilitation unit.

We found that cultures of the nares had a high nega-tive predictive value (98%) for perianal colonization, withonly 2% of patients with a negative MRSA nares culture hav-ing a positive perianal culture. As a result, we have elimi-nated routine perianal MRSA cultures for patients newly

TABLE 2FREQUENCY OF METHICILLIN-RESISTANT STAPHYLOCOCCUS AUREUS ISOLATION BY BODY SITE AND SELECTED PATIENT ATTRIBUTES

Positive MRSA CultureNares Perianal Wound � 1 Body Site

Attribute No./Total (%%) No./Total (%%) No./Total (%%) No./Total (%%)

Referral sourceSJMMC 12/295 (4) 13/295 (4) 10/64 (16) 22/295 (7)Other hospitals 12/79 (15) 4/79 (5) 10/38 (38) 18/78 (23)Skilled nursing facility 6/19 (32) 4/19 (21) 6/9 (67) 10/19 (53)Home 1/18 (6) 1/18 (6) 0/18 1/18 (6)P < .00001 .2 .007 < .00001

History of MRSAPresent 16/24 (67) 6/24 (25) 11/18 (61) 19/24 (79)Absent 15/387 (4) 16/387 (4) 16/114 (14) 32/386 (8)P < .00001 < .00001 < .00001 < .00001RR (CI95) 17.2 (9.7 to 30.5) 6.0 (2.6 to 14.0) 4.4 (2.4 to 7.8) 9.5 (6.5 to 14.1)

Prior antibioticsYes 18/173 (10) 11/173 (6) 19/71 (27) 29/173 (17)No 13/233 (6) 11/233 (5) 8/58 (14) 22/232 (9)P .1 .6 .1 .03RR (CI95) — — — 1.8 (1.1 to 3.0)

MRSA = methicillin-resistant Staphylococcus aureus; SJMMC = St. John’s Mercy Medical Center; RR = relative risk; CI95 = 95% confidence interval.

This content downloaded from 91.229.248.208 on Wed, 14 May 2014 23:47:24 PMAll use subject to JSTOR Terms and Conditions

Vol. 23 No. 9 MRSA IN AN ACUTE REHABILITATION UNIT 519

admitted to our acute rehabilitation unit. A negative MRSAnares culture also had a lower, but still relatively high, neg-ative predictive value for MRSA isolation from wound cul-tures (83%).

An analysis of patient characteristics associated withthe likelihood of a positive MRSA culture at the time ofadmission to our acute rehabilitation unit revealed that onlyprior history of MRSA infection or colonization and sourcefacility (particularly outside hospitals and skilled nursingfacilities) were independently associated with positiveMRSA screening cultures from one or more body sites.Collectively, however, these two variables accounted foronly 19 (37%) of 51 patients admitted to the acute rehabili-tation unit with MRSA colonization.

The incidence of acute rehabilitation unit–associatedMRSA infection among patients with a positive MRSAscreening culture or cultures—who routinely receiveddecolonization therapy—was relatively low (2%) and wasnot significantly different from that of patients with nega-tive screening cultures (0.8%). Because the rate of subse-quent MRSA infection among colonized rehabilitationpatients has been reported to be as high as 35% withoutdecolonization therapy,9 our attempt to eradicate coloniza-tion in newly admitted patients might have been effective inreducing the risk of subsequent MRSA infection duringtheir stay in our acute rehabilitation unit. However, proper-ly designed randomized studies are needed to further clar-ify the potential role of decolonization programs in reduc-ing MRSA infection in acute rehabilitation units.

Certain limitations of our study are worthy of discus-sion. First, the results of our study may not necessarily begeneralizable to other rehabilitation facilities whose patientprofile and average patient length of stay may be differentfrom ours. However, in a study of diagnostic coding among252 rehabilitation facilities nationwide,10 most patients didnot have diagnoses related to spinal cord injury, makingthe results of our study potentially more generalizable tothe average acute rehabilitation unit. Second, we did notperform a cost-utility analysis of routine MRSA screening

of patients newly admitted to our acute rehabilitation unit.Nevertheless, the finding of a low yield of perianal MRSAcultures in patients with negative nares cultures, and there-fore their elimination from routine screening cultures, willlikely translate into some cost savings at our unit.

As a result of our findings from the current study, wehave revised our MRSA screening policy such that allnewly admitted patients undergo nasal and wound culturesfor MRSA at the time of admission to our acute rehabilita-tion unit, regardless of the referral source or prior historyof MRSA colonization or infection. In addition, we no longerperform perianal cultures as part of our screening programfor MRSA. Further studies are needed to better define thecost-effectiveness of such an approach in reducing the riskof MRSA infection or colonization among rehabilitationpatients.

REFERENCES1. Mylotte JM, Graham R, Kahler L, Young L, Goodnough S. Epidemiology of

nosocomial infection and resistant organisms in patients admitted for thefirst time to an acute rehabilitation unit. Clin Infect Dis 2000;30:425-432.

2. Bradley SF. Methicillin-resistant Staphylococcus aureus: long-term careconcerns. Am J Med 1999;106(5A):2S-10S.

3. Maeder K, Ginunas VJ, Montgomerie JZ, Canawati HN. Methicillin-resistant Staphylococcus aureus (MRSA) colonization in patients withspinal cord injury. Paraplegia 1993;31:639-644.

4. Meyer L, Manian F, Rinke D, Laws B. Routine oxacillin-resistantStaphylococcus aureus screening of newly-admitted patients to a rehabil-itation unit. Am J Infect Control 1989;17:117. Abstract.

5. Pick FCM, Rose M, Wang D, Gardner BP, Gillett AP. The prevention ofspread of methicillin-resistant Staphylococcus aureus in a spinal injuriescenter. Paraplegia 1994;32:732-735.

6. Mylotte JM, Kahler L, Graham R, Young L, Goodnough S. Prospectivesurveillance of antibiotic-resistant organisms in patients with spinal cordinjury admitted to an acute rehabilitation unit. Am J Infect Control 2000;28:291-297.

7. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitionfor nosocomial infections, 1988. Am J Infect Control 1988;16:128-140.

8. Centers for Disease Control and Prevention. Methicillin-resistant S.aureus rates increase 260% in hospitals participating in HIP’s interna-tional monitoring systems. Focus Newsletter 2000;9(May-June):1-2.

9. Aeilts GD, Sapico FL, Canawati HN, Malik GM, Montgomerie JZ.Methicillin-resistant Staphylococcus aureus colonization and infection ina rehabilitation facility. J Clin Microbiol 1982;16:218-223.

10. Stineman MG, Escarce JJ, Tassoni CJ, Goin JE, Granger CV, WilliamsSV. Diagnostic coding and medical rehabilitation length of stay: theirrelationship. Arch Phys Med Rehabil 1998;79:241-248.

This content downloaded from 91.229.248.208 on Wed, 14 May 2014 23:47:24 PMAll use subject to JSTOR Terms and Conditions