LEEDer Group Inc.
8508 North West 66th St.
Miami, Florida 33166 USA

Phone: 305.436.5030
Fax: 305.436.0086
E-mail Address: info {at] LEEDerGroup [dot] com

Antimicrobial Resistance Long Term Care

KYDEX-PRO

STRONGER

SAFER

LEEDerGroup.com

  • LEEDer Note: all LEEDer Group products provided with Anti-Microbial and Moisture Wicking Softgoods
Antimicrobial Resistance in Long-Term–Care Facilities

Larry J. Strausbaugh, MD; Kent B. Crossley, MD; Brenda A. Nurse, MD; Lauri D. Thrupp, MD; SHEA Long-Term–Care Committee

During the last quarter century, numerous reports have indicated that antimicrobial resistance commonly is encountered in long-term–care facilities (LTCFs).

Gram-negative uropathogens resistant to penicillin, cephalosporin, aminoglycoside, or fluoroquinolone antibiotics and methicillin-resistant Staphylococcus aureus have received the greatest attention, but other reports have described the occurrence of multiply-resistant strains of Haemophilus influenzae and vancomycin-resistant enterococci (VRE) in this setting. Antimicrobial-resistant bacteria may enter LTCFs with colonized patients transferred from the hospi- tal, or they may arise in the facility as a result of mutation or gene transfer. Once present, resistant strains tend to persist and become endemic. Rapid dissemination also has been documented in some facilities.

Person-to-person transmission via the hands of healthcare workers appears to be the most important means of spread. The LTCF patients most commonly affected are those with serious underlying dis- ease, poor functional status, wounds such as pressure sores, invasive devices such as urinary catheters, and prior antimi- crobial therapy. The presence of antimicrobial-resistant pathogens in LTCFs has serious consequences not only for residents but also for LTCFs and hospitals. Experience with control strategies for antimicrobial-resistant pathogens in LTCFs is limited; however, strategies used in hospitals often are inapplicable. Six recommendations for controlling antimicrobial resistance in LTCFs are offered, and four pri- orities for future research are identified (Infect Control Hosp Epidemiol1996;17:129-140).

From the Medical Service (Dr. Strausbaugh), Veterans Affairs Medical Center and Division of Infectious Diseases, Department of Medicine, School of Medicine, Oregon Health Sciences University, Portland, Oregon; the Department of Medicine (Dr. Crossley), St Paul Ramsey Medical Center, St Paul, Minnesota, and Division of Infectious Diseases, Department of Medicine, University of Minnesota Medical School, Minneapolis, Minnesota; the Hospital for Special Care (Dr. Nurse), New Britain, Connecticut, and School of Medicine, University of Connecticut School of Medicine, Farmington, Connecticut; and the Infectious Disease Service (Dr. Thrupp), Department of Medicine, University of California–Irvine Medical Center and Department of Medicine, University of California–Irvine College of Medicine, Orange, California.*Other members of the SHEA Long-Term–Care Committee include David W. Bentley, MD; Richard A. Garibaldi, MD; Ellen G. Neuhaus, MD; Lindsay E. Nicolle, MD; and Philip W. Smith, MD. Address reprint requests to Larry J. Strausbaugh, MD, Medical Service (111F), Portland VA Medical Center, PO Box 1034, Portland, OR 97207. 95-SR-160. Strausbaugh LJ, Crossley KB, Nurse BA, Thrupp LD, SHEA Long-Term–Care Committee. Antimicrobial resistance in long-term–care facilities. Infect Control Hosp Epidemiol 1996;17:129-140.

ABSTRACT

SHEA Position Paper
Vol. 17 No. 2 INFECTION CONTROL AND HOSPITAL EPIDEMIOLOGY 129

INTRODUCTION

The purpose of this position paper is to direct attention to the problem of antimicrobial resistance in long-term–care facilities (LTCFs) and its import for patients, facilities, and the healthcare system as a whole; to recommend strategies for control; and to identify priorities for future research. The term “LTCF? refers to facilities that provide for the biopsychological needs of persons with sustained self-care deficits and includes nursing homes, chronic disease hospitals, rehabilitation centers, foster and group homes, institutions for the mental-
ly retarded, etc. In the United States, the greatest number of LTCF residents live in the estimated 20,000 nursing homes that are scattered across the country.

STATEMENT OF THE PROBLEM

Occurrence of Antimicrobial-Resistant Pathogens in LTCFs Antimicrobial-resistant bacteria in LTCFs have been documented repeatedly in the medical litera- ture since the early 1970s.

1 Information has originated from either surveillance studies of infections in LTCF residents or from outbreak investigations. While initial reports described lactam or amino glycoside-resistant gram-negative bacilli, principally isolated from the urine of catheterized residents, recent reports have highlighted other resistant pathogens, notably methicillin-resistant Staphylo- coccus aureus (MRSA). There are no comprehensive surveys, but data available from surveillance studies and focused investigations indicate that antimicrobial- resistant pathogens are encountered frequently in LTCFs. It should be noted, however, that LTCFs are affected unequally and that both prevalences and patterns of resistant pathogens vary considerably from one facility to another. Only one study has examined prospectively more than one form of resistance in a single facility. Terpenning and colleagues at the Ann Arbor Veterans Affairs (VA) Nursing Home Care Unit (NHCU) screened all residents during a 2-year period for colonization or infection with MRSA, enterococci with high-level resistance (HLR) to gentamicin, and gram-negative bacilli resistant to either gentamicin or ceftriaxone.

2 During the first year, 22.7 1% (meanSEM) of residents were colonized with MRSA, 20.2 1% of residents were colonized with enterococci expressing HLR to gentamicin, and 12.61% of residents were colonized with resistant gram-negative bacilli. The latter two rates were unchanged during the second year of the study, when MRSA colonization rates were reduced with mupirocin decolonization therapy. During the study, 49.6% of the infections with an identified etiology were caused by one of the antibiotic-resistant pathogens. Piecemeal observations from surveil- lance studies and outbreak investigations suggest that the experience at the Ann Arbor VA NHCU is not atypical.

Gram-negative uropathogens . Urine cultures from LTCF residents often yield antimicrobial- resistant pathogens. These include highly resistant strains of bacteria belonging to genera that usually are susceptible to most antimicrobial agents when isolated from the urine of nonhospitalized patients. Ampicillin-resistant isolates of Escherichia coli and cephalothin-resistant isolates of Klebsiella species, which are very common in LTCFs, exemplify this phenomenon.3-6 Other resistant urinary pathogens belong to genera that are noted for resistance to multiple agents. For example, Providencia species, which often are resistant to lactam and aminogly coside antibiotics, frequently are isolated in LTCFs from the urine of catheterized residents, but rarely are recovered in other settings. 6-10 Two studies comparing the antimicrobial susceptibility profiles of LTCF urinary isolates with those from hospital- ized patients found that the LTCF isolates generally were more resistant.3,11 Thus, gram-negative bacil-li isolated from the urine of LTCF residents demonstrate resistance to many antimicrobial agents and number among the most resistant microbes known. Aminoglycoside resistance in gram-negative uropathogens has been observed repeated- ly.2,3,5,6,11-16 For example, in a 1-day prevalence study conducted in seven nursing homes in Salt Lake City, Utah, 33% of the gram-negative uropathogens were resistant to gentamicin. 4 Similarly, in a prevalence study conducted in a VA NHCU in suburban Cleveland, gentamicin-resistant gram-negative bacilli were isolated from the urine of 16% of the residents. 15 In an in vitro susceptibility study of urine, blood, and wound isolates from residents of more than 100 nursing homes in Oklahoma, 55% of the 33 Enterobacter aerogenes strains and 61% of the 284 Pseudomonas aeruginosa strains were resistant to gentamicin. 16 Resistance to trimethoprim and the combination of trimetho- prim-sulfamethoxazole has been documented fre- quently in LTCFs. 3-6,11,16,17 Wingard et al, for example, detected colonization of urine or perineum with trimethoprim-resistant gram-negative bacilli in 52% of residents at a VA NHCU who were surveyed at quarterly intervals over a 2-year period. 17 Resistance to fluoroquinolone antimicrobials also has been reported. Prospective surveillance in seven skilled nursing facilities in southern California found that approximately one third of Pseudomonas isolates and 12% of Enterobacteriaceae recovered from urine were norfloxacin resistant.10In the study of isolates from Oklahoma nursing home residents, 50% of Paeruginosa strains, 59% of Providencia stuartii strains, and 73% of Acinetobacter calcoaceticus strains were ciprofloxacin resistant. 16 Lastly, resistance to lactam eg, penicillin and cephalosporin antibiotics has been noted commonly in LTCF uropathogens. 3-6,11,13,16 The most worrisome report of lactam antibiotic resistance described an outbreak involving 29 patients in a chronic-care facility in Massachusetts caused by ceftazidime resistant strains of Klebsiella pneumoniae and other members of the family Enterobacteriaceae.18 In the susceptibility study of isolates from Oklahoma nursing home residents, 25% of the 72 strains of Citrobacter freundii were resistant to imipenem.16 MRSA. Strains of MRSA, frequently resistant to multiple other antimicrobial agents, are the most thoroughly studied resistant pathogens in LTCFs. Following the initial report by Storch and col- leagues of an MRSA outbreak in a St Louis-area skilled-care facility, 19 the medical literature blossomed with reports of MRSA in LTCFs from all around the United States, including California, New York, Illinois, Michigan, Minnesota, Pennsylvania, Washington, Oregon, Maryland, and Kentucky. 19- 31 These reports encompass community and VA facilities, large and small facilities, urban and nonur- ban facilities, and adult and pediatric facilities. Three regional surveys in Minnesota, New York, and Oregon further attest to this phenome- non. In 1989, of 395 LTCFs in Minnesota (88%), 12% reported MRSA cases. 31 Eight percent of these facilities identified MRSA as a problem, and 69% with MRSA cases had sought outside help or con- sultation for its control. In eight counties of western New York surveyed in 1991, 81% of 75 responding facilities had identified one or more MRSA cases in the preceding year, and 21% indicated an infection control problem with MRSA. 26 Larger facilities reported more cases. Lastly, of 109 nursing homes in Oregon surveyed in 1990, only one facility had MRSA cases in 1985 and 1986, whereas 34 (31%) acknowledged MRSA cases in 1989. 27 The number of MRSA cases in the involved facilities also increased substantially during the period from 1985 to 1989. These three surveys, dispersed in time and geography, depict widespread MRSA col- onization and infection in LTCFs. The potential of MRSA strains to acquire addi- tional forms of resistance in LTCFs also has been reported. Ciprofloxacin resistance was noted in MRSA strains isolated in 1988 from residents of eight nursing homes in metropolitan New York within 3 months of the drug becoming commercially available.21 During an MRSA outbreak at a VA NHCU in Washington, MRSA strains uniformly were ciprofloxacin resistant within a year of initial ciprofloxacin use.25 Other pathogens. The prevalence of enterococci with HLR to gentamicin and other aminoglycoside antibiotics in Ann Arbor, Michigan, in 1986 was 4.3% in urine specimens and rectal swabs from residents of a community nursing home and 47.4% in those from residents of the VA NHCU. 32 The higher rate of colonization in the VA NHCU residents probably reflected the high rate of colonization (36.1%) in the acute-care division of the Ann Arbor VA Medical Center. More recently, VRE have been detected in nursing home residents. In the in vitro susceptibility study of isolates from residents of Oklahoma nurs- ing homes, vancomycin resistance was detected in 3% of Enterococcus faecalis strains and 22% of Enterococcus faecium strains.16 An outbreak caused by an ampicillin-resistant strain of H influenzaeinvolving six patients in a nursing
home and adjoining hospital during a 1-month period has been reported.33 All patients had personal contact with at least one other case-patient, suggesting person- to-person spread; two patients were bacteremic, and one died. Sturm et al reported a similar type of outbreak involving 15 patients in a pulmonary rehabilitation center in The Netherlands.34 The outbreak strain of H influenzaewas resistant to amoxicillin, trimethoprim- sulfamethoxazole, chloramphenicol, and tetracycline. Finally, Choi et al described a 1988 nursing home out- break caused by Salmonella heidelberg, a serotype frequently displaying multiple antimicrobial resistance.35

Origin of Antimicrobial-Resistant Pathogens in LTCFs

Antimicrobial-resistant pathogens in LTCFs have three possible origins. First, they may arrive with a colonized or infected patient. This has been documented for MRSA 25 and gentamicin-resistant strains of Enterobacteriaceae.36 Second, resistant pathogens may be selected for or, more rarely, may arise via mutation as a consequence of antimicrobial use for a given patient or for the facility as a whole. In their study of 10 patients with chronic indwelling uri- nary catheters at a VA NHCU in North Dakota, Bjork and colleagues observed that 70% of 63 antibiotic courses prescribed for these patients over a 2.5-year period were followed by bacteriuria with organisms resistant to the antibiotics administered. 11 Lastly, antimicrobial-resistant pathogens may arise from the transfer of genetic material from one species or genus of bacteria to another within the facility. An outbreak of ceftazidime resistance caused by an extended-spectrum lactamase in a chronic-care facility in Massachusetts arose from plasmid trans mission among different strains of the family Enterobacteriaceae and not from dissemination of a single resistant isolate following the introduction of ceftazidime into the facility.18 Similarly, in a study of gentamicin-resistant gram-negative bacterial coloniza- tion in a VA NHCU, an E coliplasmid, which conferred resistance to ampicillin, carbenicillin, tetracycline, and sulfonamides, was identical to two C freundii plasmids and a P stuartii plasmid isolated from three different patients.15Thus, all three mechanisms play a role in generating resistant bacteria in LTCFs; the relative importance of the three mechanisms has not been delineated.

Natural History and Epidemiology of Antimicrobial-Resistant Pathogens in LTCFs Once antimicrobial-resistant pathogens are introduced into LTCFs, they tend to persist and become endemic. Widespread dissemination throughout some facilities has been documented. For example, 15 months after the introduction of MRSA into a VA NHCU in Washington, a preva- lence study identified 34% of the residents and 7% of the staff to be colonized with MRSA. 25 Outbreaks of clinical disease have been observed in other LTCFs. 19 Several studies have document- ed large percentages of residents colonized by MRSA over extended periods of time. The mean monthly colonization rate over a 2-year period for MRSA at any site in a VA NHCU in Ann Arbor was 231%.23 For 65% of the 60 patients who were eval- uated for at least 3 months after initial detection of MRSA colonization, persistent or probable persis- tent colonization was documented. Similar observa- tions are reported in other studies. 24 MRSA persis- tence also has been reported from studies in com- munity LTCFs but at lower frequencies—5% to 16%.20,22 The persistence of antimicrobial-resistant bac- teria in LTCFs has been attributed to a number of factors. The presence of large numbers of residents with significant underlying diseases and indwelling foreign bodies (urinary catheters, feeding tubes, tracheostomies, etc) who frequently receive antimi- crobial therapy (see companion position paper by SHEA Long-Term–Care Committee 37) and who generally stay for months to years appears to be most important. 37,38 Additional institutional factors that may increase the likelihood of person-to-person transmission include resident interaction in two- or four-bed rooms and communal activities such as meals and various types of therapy; high patient-to- staff ratios, staffing by nonprofessional personnel, frequent turnover of staff, meager emphasis on infection control, and limited facilities for hand- washing may facilitate cross-infection. 4,38,39 Increased frequencies of residents colonized with resistant pathogens in large facilities with more skilled beds further support the importance of these factors.38 Surveys repeatedly have document- ed deficiencies in the infection control programs of US and Canadian LTCFs.40-43 Finally, antibiotic use likely contributes to the persistence of resistant strains in LTCFs. Few MRSA or other LTCF studies have focused specifically on the sources, reservoirs, and means of transmission. Nevertheless, it seems like- ly that the same factors responsible for the spread and maintenance of MRSA and other resistant pathogens in hospitals are operative. 44MRSA-infected and MRSA-colonized residents are thought to con- stitute the major reservoir and source of resistant strains. Of note, however, in the 1-year study report- ed by Bradley and coworkers at the Ann Arbor VA, only 3% of the 258 residents at risk appeared to acquire their MRSA strain from a roommate. These findings suggest that other sources may be present. Although several studies have documented MRSA nasal carriage in healthcare workers in LTCFs, none have implicated them directly in strain trans- mission.19,20,25 Several studies from LTCFs also have reported the recovery of MRSA and entero- cocci with HLR to aminoglycoside antibiotics from environmental surfaces, 23,32,45 but the potential role of the inanimate environment as a reservoir or source is otherwise unexplored. As in the hospital, person-to-person spread via direct contact, especially that between a resident and the transiently colonized hands of a healthcare worker, is thought to be the principal mode of trans- mission. Isolation of resistant pathogens from the hands of healthcare workers and observations on the timing of new cases have provided some evidence to support this theory for MRSA,46,47resistant gram- negative uropathogens,15,17 and H influenzae.33,34 As judged from one VA study, cross-colonization is a common occurrence. 17 Although not well studied, no evidence published to date has strongly implicat- ed indirect contact via environmental sources, droplet or aerosol generation by colonized patients with tracheostomies, or contaminated common vehicles as important means of spread for antimi- crobial-resistant pathogens in LTCFs.

Risk Factors for Infection and Colonization With Resistant

Pathogens in LTCFs For infection. Only two studies have identified risk factors for infection with antimicrobial-resistant pathogens in LTCFs. In a VA intermediate care ward and an NHCU, persistent MRSA colonization and dialysis were independent risk factors for MRSA infection.24 In another VA NHCU study, diabetes mellitus and peripheral vascular disease were sig- nificant independent risk factors for MRSA infec- tion2; intermittent urinary catheterization and presence of an indwelling urinary catheter were significant independent risk factors for infection caused by resistant gram-negative bacilli. 2 For colonization. A number of studies have reported risk factors for colonization with antimicrobial- resistant pathogens in LTCFs.2,15,17,20,22,23,28,38,48 These indicate that poor functional status, presence of wounds such as pressure sores, presence of invasive devices such as urinary catheters or feeding tubes, and recent antimicrobial therapy identify LTCF residents who are likely to be colonized with resistant microbes. Specific risk factors identified for the major resistant pathogen groups are listed in Table 1. Other putative risk factors that were identified only by univariate analysis also are described in several studies that employed stepwise logistic regression analy- sis.2,15,17,28,38

Consequences Resulting From Presence of Antimicrobial-Resistant Pathogens in LTCFs Antimicrobial-resistant pathogens in a LTCF have important consequences for the patients and for the facility. Facility-acquired infections with resistant organisms are the direct cause of considerable mor- bidity. There may be excess morbidity or mortality if initial empiric therapy is ineffective. Control strate- gies may limit the movement of infected or colonized residents, and, hence, their opportunities to socialize or participate in various forms of group therapy.

There are additional costs associated with more expensive antimicrobials or a need for parenteral ther- apy.48 Moreover, a small proportion of resistant iso- lates often begets overuse of new broader-spectrum antimicrobial agents, which, in turn, increases the selective pressure for resistant strains. Recent exam- ples of this phenomenon include the emergence of quinolone resistance in pseudomonads and MRSA fol- lowing increased use of ciprofloxacin for the empiric therapy of urinary tract infections possibly caused by bacteria resistant to trimethoprim-sulfamethoxazole; outbreaks caused by Enterobacterstrains resistant to third-generation cephalosporin antibiotics following overuse of ceftazidime or aztreonam; and, finally, clus- ters of VRE infections following overuse of van- comycin empirically in anticipation of possible MRSA infection or for Clostridium difficilecolitis. The presence of antimicrobial-resistant pathogens in an LTCF has additional consequences for the facility itself. As the facility recognizes increasing cases,

Vol. 17 No. 2 SHEA POSITION P APER 133
TABLE 1 RISK FACTORS FOR COLONIZATION BY ANTIMICROBIAL-RESISTANT BACTERIA IN LONG-TERM CARE FACILITIES*

Risk Factor for Colonization With: Category of Resistant Gram- Methicillin-Resistant Enterococci With High-Level Risk Factor Negative Uropathogens Staphylococcus aureus Aminoglycoside Resistance Patient characteristic Inflammatory bowel or Bedridden or chair/bed Renal failure, poor functional chronic renal disease and confined status22; status (Katz), and low serum prior pneumonia2; poor functional status23; albumin2; need for advanced nonambulatory status and male gender and urinary nursing care and bedridden/ increased length of stay14; incontinence 28 wheelchair-bound status32 decreased functional status and increased length of stay17; bladder dysfunction and advanced age38 Skin condition Presence of wounds2 Presence of wounds and Presence of wounds2 decubitus ulcers20,22; decubitus ulcers2; pressure sore present28 Invasive devices Intermittent urinary Nasogastric intubation or Intermittent urinary catheterization2; intravenous catheter 20; feeding catheterization2; urinary or urinary catheter15 tube or urinary catheter22 intravenous catheterization32 Prior antimicrobial Antimicrobial therapy in Current antibiotic therapy20 Antimicrobial therapy in therapy previous 2 weeks38 previous 3 months32 Prior colonization Prior perineal or rectal Positive MRSA culture prior colonization (a risk factor for to prevalence survey28 urinary tract colonization)15 Facility feature Residence in large facilities (219 to 330 skilled beds)38

*For studies using stepwise logistic regression or multivariate discriminant analysis techniques, putative risk factors that we re identified only by univariate analysis are not listed.