Infection Management for Geriatrics in Long Term Care Facilities

384 Strausbaugh nursing home appeared in 1970 (6); however, strains of MRSA remained uncom- mon in this setting for the next 15 years. Accordingly, this chapter largely focuses on the U.S. experience reported since 1985. II. EPIDEMIOLOGY AND CLINICAL RELEVANCE A. State and Regional Surveys Surveys conducted in Minnesota (7), western New York (8), and Oregon (9) more than a decade ago indicated that MRSA had spread to many LTCFs in those areas (Table 1). Of the 48 LTCFs—12% of respondents—reporting MRSA cases in the Minnesota survey, four indicated that it was a problem, and 33 (69%) indicated that they had sought help or consultation to manage the issue. In the New York survey, 81% of responding LTCFs acknowledged caring for an MRSA case in the previous year, and 16 (27%) of 59 facilities reporting MRSA cases acknowledged an infection control problem with this bacterium. Results of the Oregon survey offered temporal and quantitative observa- tions on the emergence of MRSA in LTCFs (9). None of the 109 reporting facili- ties had cases in 1985. One had cases in 1986, three in 1987, 11 in 1988, and 34 in 1989. During the same period, the total number of LTCF residents with MRSA in the reporting facilities increased annually from 21 in 1986 to 156 in 1989. Thus, from 1985 through 1989 both the number of facilities with MRSA cases and the total MRSA caseload in LTCFs increased steadily. Larger facilities were more likely to report MRSA cases: in 1989, 79% of LTCFs with MRSA cases had more than 50 beds. B. Frequency of MRSA Colonization Prevalence surveys that target both infected and colonized residents offer the most comprehensive assessment of MRSA infiltration into LTCFs because the ratio of Table 1 Results of Three State or Regional Surveys for MRSA in LTCFs Site Year No. LTCFs Percent Percent (reference) of survey surveyed responding reporting MRSA cases Minnesota (7) 1989 445 88 12 Eight counties in western 1990 81 93 81 New York (8) 1990 192 57 31 Oregon (9) Abbreviations: MRSA, Methicillin-resistant Staphylococcus aureus; LTCFs, Long-term care facilities.

MRSA 385 colonized residents to infected residents generally exceeds 20 to 1 (4). Counting only infected residents underestimates the magnitude of a facility’s MRSA bur- den. Colonization denotes asymptomatic persons who harbor MRSA at some body site, for example, the anterior nares (10). Detection requires bacterial cul- tures of the colonized site. In contrast, infected individuals have symptoms and signs of disease with positive cultures from the affected site. Rates of MRSA colonization in LTCFs have ranged from 5% to 34% in prevalence studies (11–21) reported from facilities in nine states (Table 2). These studies have detected nasal carriage most frequently, but rectal, perineal, wound, Table 2 Prevalence of MRSA in LTCFs* LTCF Study No. beds % Residents Comment Ref. location period (LTCF type) colonized with MRSA Nasal cultures of 74 residents; 7% of nursing home staff 11 St. Louis, MO 3/85 182 (Comm NH) 12 colonized 12 Los Angeles, 5/87 & 8/87 170 (Comm NH) 7.3 & 6.0 Nasal and wound cultures from CA all residents on two separate occasions; 3.4% and 2.3% of 13 Pittsburgh, PA 1/86–12/88 432 (VA 13.1 staff colonized NHCU) 981 total nasal cultures (obtained 14 Vancouver, 3/89 120 (VA 34 at monthly & bimonthly WA intervals); 32 residents NHCU) persistently colonized 15 Chicago, IL 8/88–11/89 150 (Comm NH) 4.9 to 15.6 Nasal and wound cultures from all residents; 7% of staff 16 Ann Arbor, MI 6/89–5/90 120 (VA 23 ϩ 1.0 colonized NHCU) (mean monthly Eight facility-wide nasal culture rate) surveys over 15 months; overall 8.7% of 994 nasal 17 Baltimore, MD 1/89–1/90 233 (Comm NH) 22 cultures positive for MRSA Monthly cultures of nose, perineum, rectum, and wounds; 25% of residents colonized on admission; only 10% of newly admitted patients acquired MRSA Cultures of nares, pressure sores, ostomy sites, urine, and sputum; 25% of new admissions in 4 months after prevalence survey found to be MRSA colonized (continued)

386 Strausbaugh Table 2 (Continued) LTCF Study No. beds % Residents Comment Ref. location period (LTCF type) colonized with MRSA 18 Ann Arbor, MI 6/89–5/91 120 (VA 22.7 ϩ 1.0 Monthly cultures of nose, NHCU) (year one) perineum, rectum, and 19 Durham, NC 12/91–1/92 wounds; mupirocin 120–125 beds 11.5 ϩ 1.8 intervention introduced in 20 Orange 1990–1992 (one VA and (year two) year two three Comm (mean County, CA (20 NHs) monthly Nasal cultures performed on all rate) consenting residents; months) 149 (Comm NH) differences between VA 27.3 (VA NHCU and three community 21 Orange 1993–1994 149 (Comm NH) NHCU) 8.1 nursing homes persisted over (3 Comm time County, CA (12 NHs) Cultures of nares and rectum 7.5 obtained quarterly over 20 (cumulative months; 4.1% of nares and during study 2.5% of rectal cultures period) positive for MRSA; 3.8% of new admissions colonized 9.7 (overall mean) Cultures of nares and rectum obtained on admission and months) quarterly; overall 35% of residents colonized at least once with Staphylococcus aureus—72% with MSSA and 25% with MRSA; 13% of carriers detected only with rectal cultures Abbreviations: VA, Veterans Affairs; NHCU, Nursing Home Care Unit; Comm, Community; NH, Nursing home; MSSA, Methicillin-susceptible S. aureus; MRSA, Methicillin-resistant S. aureus; LTCF, Long-term care facility. urine, and sputum cultures occasionally yielded MRSA. Although Veterans Af- fairs (VA) facilities often have higher colonization rates than community nursing homes, considerable overlap is noted. In the only direct comparison, MRSA col- onization rates in a VA facility were three times higher than those in three com- munity nursing homes (19). Whether this difference reflects unique features of the VA facilities, for instance, higher percentages of male residents or close affilia- tions with academic medical centers, or other factors remains unclear. Three prevalence studies reported MRSA colonization rates in LTCF healthcare workers ranging from 2.3% to 7% (Table 2). As in other healthcare set- tings, colonized workers serve as both reservoir and vector for MRSA (1–4,10).

MRSA 387 A number of other reports attest to the frequent presence of MRSA in nurs- ing homes and other kinds of LTCFs. Three observational studies in two Chicago community hospitals during the period 1984 to 1986 indicated overall that 76 (49%) of 155 patients with S. aureus isolates admitted from 25 nursing homes had MRSA (22). In contrast, only 13% of S. aureus isolates obtained from persons ad- mitted from the community were MRSA. In a study of the emergence of ciprofloxacin-resistant MRSA in New York healthcare facilities, 14 patients har- boring this strain resided in nursing homes (23). Similarly, 12 of 43 hospitalized patients from whom this strain was isolated had been admitted from nursing homes. In vitro susceptibility studies on 301 isolates of S. aureus obtained from more than 100 nursing homes in Oklahoma disclosed that 70% were resistant to methicillin (24). A microbiological survey of residents in 25 Nebraska LTCFs re- covered 91 strains of S. aureus, of which 43% were MRSA (25). A pediatric LTCF in Kentucky reported detection of 18 MRSA-colonized children during one 6-month period (26). One brief report mentions MRSA outbreaks in Canadian LTCFs (27). A retrospective study from Edinburgh indicated that 9.8% of 204 new admissions to an acute geriatric assessment and rehabilitation ward were MRSA positive (28). Finally, one older report describes MRSA colonization and infection in a rehabilitation facility. During the period October 1977 to May 1980, 84 colonizations or infections occurred in 81 residents of a 600-bed rehabilitation hospital in Los Angeles (29). C. Introduction of MRSA into LTCFs New residents who are already colonized or infected with MRSA bring this or- ganism into LTCFs at the time of their admission and serve as the initial reservoir (10,14). Asymptomatic residents transferred directly from acute care facilities where MRSA strains are prevalent probably account for most of this spread. Screening cultures in various types of facilities have indicated that 2% to 25% of new residents harbor this organism in their nose or at some other body site (15–17,20). D. Natural History of MRSA in LTCFs Once it has entered an LTCF, MRSA tends to spread and persist. Spread may be dramatic. For example, 15 months after the introduction of MRSA into a VA Nurs- ing Home Care Unit (NHCU) in Vancouver, WA, a prevalence survey indicated that 34% of residents and 7% of staff were colonized with the outbreak strain (14). A nasal prevalence study conducted almost 3 years later indicated that 10% of the facility’s residents remained colonized (30). Serial prevalence studies in other LTCFs also testify to MRSA’s persistence in this environment (13,15,16, 18,20–22). Individual residents may remain colonized for months to years.

388 Strausbaugh An outbreak of symptomatic disease has signaled the arrival of MRSA in some facilities. For example, the transfer of five patients with MRSA pneumonia during a 1-week period in 1985 brought the problem to attention in one St. Louis area nursing home (11). These cases followed in the wake of a community-wide outbreak of influenza A. A prevalence survey later indicated that 12% of the fa- cility’s residents were colonized with MRSA. E. Transmission of MRSA As in other healthcare settings, colonized or infected residents and colonized staff constitute the reservoir for MRSA (2,4,10,31–34). Person-to-person spread ac- counts for most transmission, and direct contact of residents with the hands of transiently colonized healthcare workers probably represents the principal mode of acquisition (31–34). Although uncommon, hand carriage of MRSA by health- care workers has been documented in LTCFs (35,36). Resident-to-resident trans- mission may also occur. In a 1-year prevalence study, nine residents—approxi- mately 25% of residents acquiring MRSA in the facility that year—became colonized with the same phage type as their roommate (16). In that situation, di- rect contact with the roommate or indirect contact with contaminated objects in the environment or colonized healthcare workers represents the likely means of spread. Of note, only 3% of the 258 residents at risk became colonized with MRSA during that year. Environmental contamination with MRSA has been doc- umented in LTCFs (16,34,37), but its role in transmission remains undefined. It is not known how frequently resident contacts with MRSA lead to prolonged colo- nization, but carriage rates in LTCFs suggest that it occurs commonly. F. Infection Caused by MRSA Eight studies of more than 1-year’s duration (11,13,18,19,20,21,29,38) have de- scribed 125 MRSA infections in LTCF residents (Table 3). Skin and soft tissue in- fections, urinary tract infections, and respiratory tract infections accounted for at least 46, 25, and 20 of these infections, respectively. Because these three types of infections predominate in LTCFs (39), MRSA’s involvement is not unexpected. Like methicillin-susceptible S. aureus (MSSA) in LTCFs, MRSA causes skin and soft tissue infections with greater frequency than any other types of infections (21,38). In the eight studies (Table 3), bacteremia complicated at least nine infec- tions, and at least four patients succumbed to their infections. The care of at least 27 of these patients required transfers to hospital. Generally, MRSA infections arise in residents who have been colonized for various lengths of time. The risk of colonized residents developing infection varies considerably and depends to some extent on comorbid conditions such as

MRSA 389 Table 3 MRSA Infections in LTCFs Facility Total no. Types of Comments on (mos of MRSA MRSA infections rates and Ref. study) infections SSTI UTI RTI Other complications 29 Rehab. 28 15 11 0 2 Prospective surveillance Hosp (32) 3 2 12 on all wards; 56 colonized residents also 11 Com NH 17 Not stated detected; one bacteremia (13) 0 Retrospective and 13 VA NHCU 15 prospective (24) VA surveillance; 5 INCU pneumonias associated (30) with influenza outbreak; no bacteremias; one 38 VA NHCU 28 15 5 5 death (60) 25% of persistent carriers of MRSA carriers had episode of staphylococcal infections, compared with 4% of persistent MSSA carriers and 4.5% of noncarriers; rate of development of infection among MRSA carriers: 15% for every 100 days of carriage; high percentage of infections in dialysis patients 3 Incidence of MRSA infection ranged from 0.07 to 0.32 cases per 1000 resident care days; incidence of MSSA infection ranged from 0.15 to 0.29 cases per 1000 resident care days; 3 bacteremias; 12 MRSA infections required transfer to hospital; 3 MRSA infections were fatal (continued)

390 Strausbaugh Table 3 (Continued) Facility Total no. Types of Comments on (mos of MRSA MRSA infections rates and Ref. study) infections SSTI UTI RTI Other complications 18 VA NHCU 15 8 NS NS 7 MRSA carriage rates (24) averaged 11% to 22% during study period; 19 VA NHCU 8 MRSA accounted for and 3 less than 5% of LTCF Com NHs infections; mupirocin (12) intervention in year 2; 2 bacteremias; 10 20 Com NH 12 hospitalizations; no (20) deaths 21 Com NH 14 1 2 1 4 3 bacteremias; risk of (12) infection 6.4 times higher in those previously colonized with MRSA; rates of infection not different between VA and 3 community nursing homes 5 6 1 0 7% of 15 colonized newly admitted residents and 5% of in house colonized residents developed infection 4 5 2 3 No hospitalizations; bacteremias not mentioned; 15 MSSA infections during study period; patients colonized with MRSA not more likely to develop infection than MSSA colonized residents Abbreviations: MRSA, Methicillin-resistant Staphylococcus aureus; LTCF, Long-term care facility; SSTI, Skin and soft tissue infection; UTI, Urinary tract infection; RTI, Respiratory tract infection; Com NH, Community nursing home; VA, Veterans Affairs; NHCU, Nursing home care unit; INCU, Intermediate care unit; MSSA, Methicillin-sensitive Staphylococcus aureus.

MRSA 391 pressure ulcers and influenza. Three studies in VA facilities have addressed this risk in different ways. In one report, the highest estimates of infection in MRSA- colonized residents found that 25% of 32 persistently colonized residents devel- oped staphylococcal infections (13). In contrast, only 4% of residents persistently colonized with MSSA and 4.5% of residents not colonized with S. aureus became infected. Thus, in this study, persistent colonization with MRSA carried a signif- icantly greater risk of subsequent infection. The high frequency of infection in res- idents in the intermediate nursing care unit who had higher rates of chronic ob- structive pulmonary disease and chronic renal failure requiring dialysis may have accounted for the higher rates of infection (40). Other VA studies have reported much lower percentages of infection in MRSA-colonized individuals. For example, in a 1-year study, only 3% of 341 pa- tients at risk developed MRSA infection even though MRSA carriage rates ex- ceeded 20% (16). In a 5-year study, the overall rate for S. aureus infections re- mained fairly stable after the introduction of MRSA into the facility during year 2, even though 34% of residents became colonized with MRSA (38). Annual rates for all S. aureus infections ranged from 0.29 to 0.47 infections per 1000 resident care days over the entire period. The percentage of infections caused by MRSA increased over time, but the overall percentage of infections caused by S. aureus remained in the narrow 13% to 17% range. Moreover, infection rates for the en- tire facility remained steady. In another recent study, the risk of infection in MRSA-colonized residents in one VA facility and three community nursing homes was 6.4 times (95% con- fidence interval (CI), 2.3 to 18.0) greater in residents colonized at baseline (19). No statistically significant increased risk in the VA facility was detected. How- ever, the risk of infection in MRSA-colonized residents in the three community nursing homes was 15 times (95% CI, 13.3 to 73.3) that seen in noncolonized res- idents. The rates of MRSA infection in the VA facility and three community nurs- ing homes did not differ significantly in the 1-year of study, even though colo- nization rates were higher in the former. Of note, the 0.16 and 0.12 per 1000 resident care-day MRSA infection rates observed in the VA and three community nursing homes, respectively, were similar in magnitude to those previously re- ported (38). Finally, a study of both MSSA and MRSA infections in a community skilled nursing facility over the course of a year noted that MRSA infections arose in previously colonized individuals at the same rate that MSSA infections arose in previously colonized individuals (21). In this study, MRSA and MSSA infection rates approximated one another—0.27 and 0.29 infections per 1000 resident care days, respectively. In sum, MRSA infections usually arise in colonized residents, but only a small percentage of colonized residents become infected. Infections caused by MRSA in LTCFs are similar to those caused by MSSA in this environment. Un- like other common bacterial pathogens in LTCFs, they cause skin and soft tissue infections with greater frequency than urinary or respiratory tract infections.

392 Strausbaugh G. Risk Factors for Colonization and Infection by MRSA in LTCFs Risk factors for MRSA colonization and infection in LTCFs mirror those asso- ciated with other antimicrobial-resistant bacteria (10). In general terms, these in- clude poor functional status, conditions that cause skin breakdown, presence of invasive devices, prior antimicrobial therapy, and a history of antecedent colo- nization. Studies using multivariate analysis have identified the following spe- cific risk factors for MRSA colonization in LTCFs: male gender (17); urinary incontinence (17); fecal incontinence (19); presence of wounds (18), pressure ul- cers (17–19); nasogastric intubation (12); antibiotic therapy (12); and hospital- ization within previous 6 months (19). Studies using only univariate analysis have identified similar putative risk factors for MRSA colonization in LTCFs: bedridden or chair/bed confined status (15), poor functional status (16), pressure ulcers (15), feeding tubes (15), urinary catheters (15), prior LTCF-associated in- fections (20), and therapy with nonquinolone antimicrobial agent in prior 3 months (20). Only two VA studies have examined risk factors for MRSA infection in LTCFs. Using stepwise logistic regression analysis, one study found persistent MRSA colonization and dialysis to be significantly associated with infection with odds ratios (ORs) of 5.9 (95% CI, 2.2–15) and 4.7 (95% CI, 1.8–12), respectively. Another study used similar methods and identified diabetes mellitus (OR ϭ 5.1; 95% CI, 2.1–18.6) and peripheral vascular disease (OR ϭ 4.3; 95% CI, 1.3–14.3) as risk factors for MRSA infection (18). III. CLINICAL MANIFESTATIONS A. Syndromes and Pathogenesis Methicillin-resistant Staphylococcus aureus and MSSA appear to have equiva- lent virulence for humans (1–4). Accordingly, the kinds of infections caused by MRSA and their clinical features are virtually identical to those caused by MSSA. In the 5-year experience of one VA facility (38), there was no signifi- cant correlation between the site of infection and the methicillin susceptibility of the infecting strain of S. aureus. In that study, which compared all MRSA and MSSA infections, skin and soft tissue infections (including conjunctivitis and otitis externa) accounted for 38% of all staphylococcal infections. Pneumonia accounted for 30% of staphylococcal infections, and urinary tract infections for 25%. Staphylococcus aureus on cutaneous surfaces combined with breeches in the integrity of skin and mucous membranes likely give rise to skin and soft tissue infections, which may include: cellulitis, surgical site and other wound infections,

MRSA 393 bursitis, perianal and skin abscesses, infected pressure sores, infected leg ulcers, and paronychia (38). Nasal colonization in association with aspiration probably contributes to the development of pneumonia. Staphylococcus aureus on perineal skin or genital membranes and the presence of indwelling urinary catheters likely predisposes LTCF residents to staphylococcal urinary tract infections. Both MRSA and MSSA can invade the bloodstream and give rise to distant site infections such as arthritis, endocarditis, osteomyelitis, visceral abscesses, and others. Bacteremias and distant site infections account for a small percentage of staphylococcal infections in LTCF residents (18,38). In some cases, bone and joint involvement may result from local invasion, for example contiguous spread from infected pressure ulcers. B. Clinical Features The manifestations of various infectious syndromes caused by MRSA in LTCF residents are similar to those caused by other pyogenic bacteria. For example, res- idents with cutaneous abscesses usually exhibit fever and varying degrees of red- ness, swelling, warmth, and tenderness surrounding the abscess. Residents with pneumonia generally manifest fever in association with respiratory symptoms such as cough and shortness of breath. Likewise, residents with MRSA urinary tract infections characteristically exhibit fever and local symptoms such as dy- suria, frequency, urgency, and suprapubic pain. Residents with MRSA conjunc- tivitis exhibit inflamed conjunctivae and purulent discharge (41). Notwithstanding these generalities, elderly nursing home residents often have atypical presentations for MRSA infections as they do for those caused by other etiological agents. Local and systemic inflammatory response may be di- minished, resulting in decreased temperature elevations and blunting of local manifestations (39,42). Neurological deficits and cognitive impairment, which are common in elderly nursing home residents, may also obscure symptoms and signs of MRSA and other types of infection. Accordingly, LTCF practitioners maintain a high index of suspicion for infection and subtle signs, such as minor changes in mental or functional status, as possible indicators of MRSA and other types of in- fections. IV. DIAGNOSTIC APPROACH A. Recognition and Delineation of Clinical Syndrome Methicillin-resistant Staphylococcus aureus-infected LTCF residents come to clinical attention in the usual ways. Reports from nursing staff about temperature elevation or other alteration in vital signs often prompt evaluation, as do those that

394 Strausbaugh describe specific symptoms or signs of infection. Diminished cognitive function and inability to perform usual activities of daily living (ADLs) also bring patients to clinical attention (42). Often the resident’s history and a limited physical ex- amination disclose the presence of MRSA infection. For example, residents with a new cough, tachypnea, and rales over one lung field probably have pneumonia. Pneumonia caused by MRSA enters the differential diagnosis from the outset in known carriers and in facilities with high rates of colonization or infection. Simi- larly, the resident with a colonized pressure ulcer who develops fever and redness, swelling, and tenderness extending out from the margins of the ulcer likely has an MRSA secondary infection of that site. The use of laboratory tests, radiography, and other ancillary procedures in the diagnosis of MRSA infections conforms to current guidelines (42). Leukocy- tosis, infiltrates on chest radiographs, pyuria, and bacteriuria (especially in un- catheterized residents), and meaningful Gram stain results from respiratory secre- tions or cutaneous exudates, all help to define syndromic diagnoses. B. Etiological Diagnosis On Gram-stained smears, both MSSA and MRSA appear as gram-positive cocci, often in clumps, or grape-like clusters. Gram-stain findings do not distinguish the two. Strains of both bacteria grow easily on most nonselective media, such as, blood agar, yielding white to yellowish colonies within a day or less (1). Rapid tests for coagulase production readily distinguish S. aureus from other species of Staphylococcus (43). Distinguishing MRSA from MSSA usually requires antimi- crobial susceptibility tests, which typically necessitate a second day for comple- tion. Therefore, isolation of MRSA strains generally requires 36 to 48 hours. Some clinical laboratories offer faster service by identifying MRSA strains with gene probes that detect mecA (43). The etiologic diagnosis awaits finalization of culture results and their inter- pretation in the context of the resident’s illness and course. Isolation of MRSA from the blood cultures of symptomatic residents virtually always indicates MRSA in- fection, whereas isolation from respiratory secretions, cutaneous exudates, and urine require interpretation to distinguish colonization from infection. In residents with strong clinical evidence for a specific infectious syndrome, isolation of MRSA in pure culture often solidifies the etiologic diagnosis, especially when Gram-stain results indicate that the bacterium is present in large numbers. Isolation of MRSA with other potential pathogens in the same culture and isolation of MRSA from a site not clearly involved by an infectious process, for example, from urine in a catheterized resident with no genitourinary symptoms, provoke the greatest diag- nostic uncertainty. Nevertheless, from a therapeutic standpoint, few practitioners can dismiss such isolates obtained from symptomatic residents because MRSA may be an etiological participant.

MRSA 395 V. THERAPEUTIC INTERVENTIONS A. Vancomycin The glycopeptide antibiotic, vancomycin, is the drug of choice for serious MRSA infections. Most strains of MRSA and MSSA are inhibited by concentrations less than 4 ␮g/ml. A few reports have described strains of MRSA with decreased sus- ceptibility to vancomycin (MICs Ն4 but Յ16 ␮g/ml), but they remain rare (44). In patients with normal renal function, vancomycin is administered intravenously in a dose of 1.0 g every 12 hours (45). Patients with renal insufficiency require dosage modifications. Many clinicians measure peak and trough serum concen- trations at least once during therapy, aiming to keep peak concentrations less than 50 ␮g/ml and trough concentrations in the 5 to 10 ␮g/ml range. Treatment courses for most MRSA infections generally range from 10 to 14 days; however, endo- carditis and osteomyelitis necessitate treatment courses of 4 to 6 weeks’ duration. In the past, the need for intravenous vancomycin therapy and monitoring ne- cessitated transfer of MRSA-infected residents to hospital. Now, many nursing homes can manage these requirements. Adverse reactions include fever, chills, and phlebitis at the infusion site, which slow infusion rates may prevent. Slow in- fusion rates also prevent the “red-man” syndrome. Rash, other allergic manifesta- tions, leukopenia, thrombocytopenia, and eosinophilia are occasionally observed but resolve when vancomycin is discontinued. Ototoxicity, the most worrisome side effect, occurs infrequently when serum concentrations are kept below 30 ␮g/ml (45). In the past, vancomycin was thought to be highly nephrotoxic. This is not the case with current preparations, but vancomycin can potentiate the nephro- toxicity of aminoglycoside antibiotics when used concurrently with them. B. Linezolid Since 2000, linezolid, an oxazolidinone derivative, has offered an alternative to vancomycin for MRSA infections. Concentrations of 4 ␮g/ml inhibit most clinical isolates of MRSA (46). Oral or intravenous doses of 400 mg or 600 mg adminis- tered every 12 hours produce serum concentrations that range from 4 to 25 ␮g/ml throughout the dosing interval (47). Orally administered preparations offer an ap- parent advantage for treating infections in LTCFs, but experience is limited. To date, few adverse reactions have complicated linezolid therapy. Diarrhea, nausea, and headache have occurred in less than 3% of recipients (46). Leukopenia, ab- normal liver function tests, and rash have occurred less frequently (see Chapter 11). Limited clinical observations have indicated that linezolid offers effective therapy for staphylococcal and enterococcal infections (46,48). In a randomized, double-blind, controlled trial conducted in patients with nosocomial pneumonia, results with linezolid compared favorably to those with vancomycin (49). Specif- ically, linezolid therapy eradicated MRSA in 15 (65%) of 23 infections and van-

396 Strausbaugh comycin eradicated MRSA in 7 (78%) of 9 infections. Notwithstanding these re- sults, linezolid’s limited track record dictates that vancomycin remain the drug of choice. However, linezolid has considerable promise. Its use will increase as op- timal circumstances and indications for treatment of MRSA infections in LTCFs become delineated. C. Other Systemic Antimicrobial Agents In vitro susceptibility tests sometimes suggest that cephalosporin antibiotics pos- sess inhibitory activity against MRSA, but clinical failures have attended their use (50). They should not be used. Ciprofloxacin once appeared to have therapeutic promise for MRSA; however, resistance has become widespread (37). The com- bination of quinupristin and dalfopristin, which is marketed as Synercid®, may offer another therapeutic alternative for MRSA infections, but experience is lim- ited (45). Strains of MRSA possess resistance to virtually all other antimicrobial agents except rifampin, trimethoprim-sulfamethoxazole, minocycline, and clin- damycin. These agents have occasionally been used in efforts to eradicate colo- nization, and in the treatment of less serious MRSA infections or in follow-up to vancomycin therapy when oral therapy was desired (31,32,37,51). Such uses have been successful, but resistance can emerge rapidly, and this problem has greatly limited their use for MRSA. D. Topical Agents For the most part, topical agents have no role in the treatment of MRSA infections. Occasionally, however, patients with superficial MRSA skin infections who lack systemic symptoms or signs may benefit from topical therapy with mupirocin oint- ment (52). Mupirocin also has been used to eradicate colonization of residents in LTCFs and other settings (31,32,52–55). Over the years, therapeutic efforts to erad- icate MRSA nasal colonization in asymptomatic carriers—decolonization ther- apy—have involved other topical agents, such as bacitracin ointment, neomycin, vancomycin, and gentamicin, but none have proved consistently efficacious (31,52). Similarly, efforts to eliminate MRSA from the skin have included skin cleaning agents containing chlorhexidine, hexachlorophene, triclosan, and povi- done iodine (31,55). Their efficacy in decolonization regimens remains unproven. E. Role of Drainage, Debridement, and Other Surgical Procedures Cure of MRSA and MSSA infections associated with abscesses, devitalized tis- sue, and closed spaces, such as, joints or pleural cavity, usually requires drainage

MRSA 397 or debridement (1). Abscesses, depending on their location and size, require either drainage from percutaneously placed catheters or needles or drainage from an open surgical procedure. Repeated needle aspirations generally suffice for in- fected joints, except for the hip, which requires open surgical drainage. Pleural empyemas require chest tube thoracostomies and, rarely, decortication proce- dures. Surgical debridement is necessary to cure chronic osteomyelitis or os- teomyelitis associated with peripheral vascular disease. Also, MRSA-infected arthroplasties and other infections involving prosthetic material generally neces- sitate removal of foreign material and surgical debridement. Management of en- docarditis may require valve replacement surgery. It is apparent that management of these infections will generally require transfer to an acute care facility. VI. INFECTION CONTROL MEASURES A. General Considerations In LTCFs, opinions about appropriate measures for controlling MRSA run the gamut from those favoring do-nothing, laissez-faire approaches on one extreme to those favoring do-everything, hospital-like approaches on the other. Unfortu- nately, there are virtually no controlled trials of different strategies to focus the dis- cussion or to inform the development of policy (see Chapters 8, 9, and 10). Never- theless, the last decade has witnessed the emergence of consensus on key principles for management of MRSA in LTCFs (Table 4). Some areas of controversy persist, but there is general agreement on the following points (10,31,33,34,56–63): 1. Virtually all LTCFs can provide good care for MRSA-colonized and - infected residents without jeopardizing the well being of other residents. In a re- view of the literature, the author noted, “In five nursing homes where MRSA was endemic, 95 infections with 5 deaths occurred during 12 years of surveillance with 12,000 admissions” (34). Others have made similar observations regarding the safety of caring for MRSA-positive residents (64). Efforts to restrict admission of colonized or infected residents usually fail because detection of carriage can be difficult. In one LTCF study, nasal cultures failed to identify 13% of MRSA car- riers (21). Restricting or delaying transfers also imposes an unnecessary burden on other sectors of the healthcare system (65). There is no evidence to suggest that screening potential admissions for MRSA and decolonizing those who are posi- tive reduces LTCF colonization or infection rates, and this approach is not rec- ommended (10). 2. LTCFs are not hospitals. Few facilities have more than a few private rooms for isolation. Few have laboratory resources necessary for screening. Re- habilitation and socialization needs of residents and communal activities, such as eating in dining rooms, limit use of isolation and stringent barrier precautions that are often used in hospitals. Moreover, some control measures may affect resi-

398 Strausbaugh Table 4 Infection Control Measures for Management of MRSA Colonization and Infection in LTCFs Endemic situation—few infections Outbreak or high endemic infection rate Surveillance Consultation From microbiology reports on With experienced epidemiologist established residents from local/regional hospital or From hospital records of new state/local health department residents or returing transfers Consultant to advice on use of Establish baseline rates of measures below colonization and infection Enhanced surveillance Education and communication Consider screening cultures of Create awareness and alleviate fear residents or staff of MRSA Consider typing MRSA isolates Emphasize importance of Standard Precautions Patient placement Consider using private rooms for Use of antimicrobial agents MRSA cases Avoid unnecessary usage Consider cohorting MRSA-positive Monitor for appropriateness residents and staff Otherwise place MRSA cases in Precautions rooms with residents who lack risk Standard Precautions for most factors for colonization residents Contact Precautions for residents Other measures whose drainage or respiratory Consider greater use of Contact secretions cannot be contained Precautions Consider (rarely) decolonization therapy Abbreviations: MRSA, Methicillin-resistant Staphylococcus aureus; LTCF, Long-term care facility. dents’ quality of life adversely (66). Rapid discharge of colonized residents is sel- dom possible. Accordingly, control strategies in LTCFs necessarily differ from those used in hospitals (10,34). 3. Prudent use of antimicrobial agents by providers plays a key role in fa- cility management of MRSA and other antimicrobial resistant pathogens (67,68). 4. Once in LTCFs, MRSA will likely persist. Aggressive approaches after MRSA’s first appearance occasionally drive it out (69); however this result is the exception to the rule (37). 5. Long-term care facilities need to perform enough surveillance to deter- mine their status with regard to MRSA and other antimicrobial-resistant pathogens.

MRSA 399 6. Judicious uses of the limited infection control resources in LTCFs ne- cessitate distinguishing between endemic and epidemic MRSA situations as well as between MRSA colonization and infection. 7. Long-term care facility settings with predominantly endemic cases of MRSA colonization primarily require appropriate use of Standard Precautions. 8. Long-term care facility settings with MRSA outbreaks, especially those with substantial morbidity due to MRSA infections, require more stringent infec- tion control measures in addition to Standard Precautions. B. Surveillance Knowledge of MRSA presence and prevalence requires some level of surveillance in LTCFs (see Chapter 9). For most facilities, regular scrutiny of microbiology re- ports and review of discharge summaries or other records for new admissions and returning transfers will suffice in nonoutbreak settings. Ordinarily, this activity falls within the purview of the infection control program, which coordinates or performs data collection and maintains records with a frequency relevant to the magnitude of the problem (70). Long-term care facilities can use this information to establish their baseline, which permits identification of outbreaks and informs decisions about control measures. Classifying MRSA cases as colonized or in- fected enhances all descriptions of a facility’s experience (10). Some workers in this field have advocated routine cultures of all new ad- missions to identify MRSA carriers (60,63), whereas others question the utility of this practice in nonoutbreak settings. Few LTCFs have the resources to per- form this task. Because identification of all MRSA carriers requires multiple cultures from different sites, including the rectum, a universal screening policy is generally regarded as onerous. Finally, screening only makes sense if it dic- tates changes in management for MRSA-positive residents, and in most nonout- break settings it does not alter room assignments, precautions, or medications. In their position paper, the Society for Healthcare Epidemiology (SHEA) Long- Term Care Committee specifically recommends against this practice in nonout- break settings (10). C. Education and Communication In all LTCFs with MRSA or other antimicrobial-resistant pathogens, education of staff and, to some extent, residents alleviates fear about these organisms and fa- cilitates appropriate management of colonized or infected residents (10,56–63). Periodic updates to LTCF staff using recent surveillance data help to create and maintain awareness of the issue. They may stimulate or rekindle the desire to use precautions appropriately.

400 Strausbaugh D. Antibiotic Use Physicians and other providers use a large amount of antimicrobial agents in LTCFs, and a number of surveys have indicated that much of this use is inappro- priate (67,68) (see Chapter 11). Because antibiotic use predisposes to MRSA col- onization (12), reducing inappropriate use may offer benefit on the individual level. It may also offer benefit on the facility level, if reducing the number of col- onized residents leads to use of more narrow-spectrum agents, lessening selective pressure that favors the emergence or persistence of more resistant pathogens like MRSA. In recent years several groups have offered comprehensive guidelines for use of antimicrobial agents in LTCF residents, hoping to reduce inappropriate use and, possibly, selection pressure (67,71). E. Precautions Standard Precautions, which combine elements of Universal Precautions and Body Substance Isolation, entered the world of medicine with publication of the 1996 Guideline for Isolation Precautions in Hospitals by the Hospital Infection Control Practices Advisory Committee (72) (see Chapter 8). Standard Precautions embodies the concept that all patients and all patient specimens should be handled as if they were infectious, capable of transmitting disease. They would seem ideal for prevention of MRSA transmission, which almost exclusively involves person- to-person spread by direct contact and often involves contact between healthcare workers and asymptomatic carriers (31–33). They emphasize hand washing after direct contact with patients and potentially infectious material, especially between contacts with different patients. Standard Precautions also dictates use of gloves, masks, eye protection, and gowns when necessary to prevent contact between in- fectious material and the healthcare worker. When used appropriately and consis- tently, these measures should interrupt transmission from one resident to another by the transiently contaminated hands of healthcare workers. The additional value of using antimicrobial soaps remains unclear (31). Hand-cleansing agents offer an alternative to soap and water (73). The position paper from the SHEA Long-Term-Care Committee (10) rec- ommends that “Routine precautions in LTCFs include adequate sinks, education, and incentives to ensure good hand-washing practices throughout the facility at all times . . . and adequate supplies and education to ensure that appropriate barrier precautions are used in the management of all wounds and invasive devices.” At- tention to these considerations facilitates the use of Standard Precautions in LTCFs. In nonoutbreak settings, most residents colonized or infected with MRSA do not require use of additional precautions in their care. Moreover, as long as they do not have large wounds or other lesions that cannot be contained by dressings or tracheostomies with excessive secretions, most authorities would not limit their movement within the LTCF or their participation in LTCF activities (10). Never-

MRSA 401 theless, residents known to be colonized or infected with MRSA should not be placed in rooms with debilitated, nonambulatory residents, that is, those at great- est risk for subsequent colonization and infection. Residents with large wounds or draining lesions that cannot be contained and those with tracheostomies and diffi- culty handling secretions generally require a higher level of scrutiny and, often, an additional layer of precautions (10). If such residents can be linked epidemiolog- ically to MRSA infection in other residents, then placing them in a private room or cohorting them with similar residents is prudent, as is restriction of their move- ment and participation in group events. In addition, Contact Precautions, which require gowns and gloves for all persons entering the room, as well as handwash- ing after glove removal, should be strongly considered (72). F. Outbreak Management Issues 1. Definition Fundamentally, an outbreak represents an increase in caseload that exceeds the baseline rate. The more accurate baselines reflect several years of experience and delineate an expected range of random variation. The SHEA position paper advo- cates defining outbreaks in terms of infections, not colonization (10). As exam- ples, it suggests that more than three infections in a week or twice the number of infections in a month than had been observed in each of the three preceding months qualify as an outbreak. Lastly, this paper suggests that situations with high endemic rates of infection, which it defines as more than one infection per 1000 resident care days, be treated like outbreaks (see Chapter 10). 2. Consultation Once an MRSA outbreak or high endemic rate of infection is recognized, the SHEA position paper recommends consultation with an experienced epidemiolo- gist. Hospital epidemiologists at local or regional hospitals, senior infection con- trol practitioners, state or local health officials, and others may qualify for this role, especially if they are knowledgeable about infection control issues in LTCFs. Consulting epidemiologists can offer independent confirmation of the problem, provide an analysis of possible causes, and offer potential solutions. Ideally, they customize their approaches to the specific circumstances and needs of a given fa- cility. As a rule, their judgments will dictate consideration of enhanced surveil- lance, additional isolation precautions, and decolonization efforts. 3. Enhanced Surveillance Outbreaks and high endemic rates of infection usually precipitate some discussion about culturing new admissions, established residents, or staff to identify asymp- tomatic carriers who might be playing a pivotal role in transmission. Costs and un-

402 Strausbaugh certainty about management of identified carriers generally discourage such screening, except in the presence of severe and protracted outbreaks. Typing of MRSA strains can solidify epidemiological links between cases and generate hypotheses about transmission. Investigations of hospital outbreaks frequently involves molecular typing methods (2); investigations of a few LTCF outbreaks also have used them (14,16,20). Cost, availability, and time issues pre- clude their use in most LTCF settings. Of note, antibiograms perform poorly in comparison to molecular typing methods (2,74). 4. Isolation and Cohorting In the setting of outbreaks and high endemic rates of infection, segregation of MRSA-colonized and -infected residents may diminish transmission (10,56– 58,61,62) Depending on the facility layout, segregation could involve use of sin- gle rooms for MRSA-colonized or -infected individuals, especially for those linked epidemiologically to other cases and those likely shedding large numbers of bacteria (from large, uncovered wounds, for example). Although disruptive, co- horting MRSA-colonized and -infected residents and, possibly, colonized staff may protect susceptible residents from additional exposure. When private rooms and cohorting fail to provide adequate segregation, placing MRSA cases in rooms occupied by healthier individuals without risk factors for colonization or infection may limit transmission. Control of outbreaks and reduction of high endemic rates may also require limiting admissions, restricting movement of MRSA-positive residents, and se- lective use of Contact Precautions (72). Because these actions disrupt the func- tioning of most LTCFs and cause considerable hardship for residents, their use re- quires sufficient provocation and justification. Individual facilities should modify or adjust the use of such measures to their specific circumstances. 5. Decolonization Because a large percentage of MRSA infections arise in colonized individuals (13,75,76), various investigators have attempted to eradicate the carrier state with antimicrobial therapy. If successful, this therapy would reduce an individual’s risk of infection and diminish a facility’s reservoir of MRSA. Unfortunately, when used to quell outbreaks or reduce high endemic rates of colonization in LTCFs or hospitals, the combined use of several different control measures has obscured evaluation of decolonization therapy, per se (2,10,11,18,26,31,37,54,69). Conse- quently, the concept of decolonization lacks supporting evidence of efficacy. There are several other problematic considerations. First, decolonization is not always successful; it frequently fails in debilitated patients with significant un- derlying disease, especially in those with open wounds or invasive devices (2,37,51,53–55). Paradoxically, decolonization often fails in those who have the

MRSA 403 greatest risk of infection. Second, use of various agents in decolonization regi- mens invariably induces resistance to the agents used. For example, in one study using rifampin-containing regimens, rifampin-resistant isolates were recovered from 80% of the 20 residents who remained persistently colonized or became re- colonized with MRSA during the 30-day follow-up period (37). Likewise, during a 7-month mupirocin intervention trial in one facility, mupirocin-resistant MRSA was isolated from 10.8% of residents (54). Finally, decolonization entails considerable expense, and it exposes resi- dents to the various toxicities of the agents used. For these reasons, routine use of decolonization therapy is not recommended in healthcare settings (32,33). Long- term care facilities should consider this strategy only in the setting of an outbreak associated with substantial morbidity, and even then, with careful monitoring by an experienced epidemiologist. In the rare circumstance when an LTCF uses a decolonization strategy, mupirocin would probably be the agent of choice. Topical application of 2% mupirocin ointment to nares for 5 days and to colonized cutaneous sites for 2 weeks will eradicate colonization in 90% of residents (1,18,52–55). However, coloniza- tion commonly recurs in 20% to 30% of residents during the weeks and months that follow treatment. Orally administered antimicrobial regimens for decolonization usually contain rifampin with or without one or two other agents (1,2,37,51). Af- ter a week of such therapy, follow-up cultures are negative in 60% to 90% of re- cipients. More than half will become recolonized in the weeks and months that fol- low. Therefore, decolonization therapy is effective in the short run, a period of 1 to 2 weeks. For a sizeable percentage of residents, however, the effect is not sustained, and resistance to the agent used appears in isolates obtained subsequently. VII. PREVENTION No single measure can prevent MRSA colonization or infection. However, atten- tion to several basic principles will likely minimize acquisition by uncolonized residents. Prudent use of antimicrobial therapy, avoidance of invasive devices, such as nasogastric tubes, and efforts to prevent pressure ulcers will probably lower an individual’s risk for colonization. Consistent use of Standard Precautions and Contact Precautions, when indicated, will interrupt the cycle of transmission. All of these efforts require a knowledgeable and compliant staff, underscoring the need for education, communication, and feedback in the infection control pro- gram. Surveillance activity helps to maintain awareness and serves to identify trends that may require additional attention. Although controversial, on occasion an elective surgical procedure on an LTCF resident may prompt consideration of preoperative decolonization and pro- phylactic antimicrobial therapy with vancomycin (77,78). For example, known

404 Strausbaugh MRSA carriers scheduled for total hip arthroplasty may have reduced risks of post- operative surgical site infections if they receive decolonization therapy preopera- tively. This same possibility applies to MSSA-colonized residents, and the results of a trial that used preoperative therapy with mupirocin to eradicate staphy- lococcal carriage are eagerly awaited. No formal recommendation currently sup- ports its use (78). Using vancomycin instead of a first-generation cephalosporin an- tibiotic for perioperative prophylaxis may also reduce postoperative MRSA infection rates, but this approach is generally reserved for hospitals with high rates of MRSA surgical site infections (77,78). Its routine use is not recommended (78). Both the preoperative decolonization and vancomycin prophylaxis strategies await additional evidence of benefit before they can receive a firm endorsement for use as a preventive measure in LTCF residents undergoing elective surgical procedures. REFERENCES 1. Waldvogel FA. Staphylococcus aureus (including staphylococcal toxic shock). In: Mandell GL, Bennett JE, Dolin R (eds). Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 5th ed. Philadelphia, W.B. Saunders Company, 2000:2069–2092. 2. Hartstein AI, Mulligan ME. Methicillin-resistant Staphylococcus aureus. In: Mayhall CG (ed). Hospital Epidemiology and Infection Control, 2nd ed. Philadelphia, Lippin- cott Williams & Wilkins, 1999:347–364. 3. Brumfitt W, Hamilton-Miller J. Methicillin-resistant Staphylococcus aureus. N Engl J Med 1989; 320:1188–1196. 4. Bradley SF. Methicillin-resistant Staphylococcus aureus infection. Clin Geriatr Med 1992; 8:853–868. 5. Centers for Disease Control and Prevention NNIS System. National Nosocomial In- fections Surveillance (NNIS) system report, data summary from January 1992-April 2000, issued June 2000. AJIC Am J Infect Control 2000; 28:429–448. 6. O’Toole RD, Drew WL, Dahlgren BJ, Beaty HN. An outbreak of methicillin-resis- tant Staphylococcus aureus infection: Observations in hospital and nursing home. JAMA 1970; 213:257–263. 7. Thurn JR, Belongia EA, Crossley K. Methicillin-resistant Staphylococcus aureus in Minnesota nursing homes. J Am Geriatr Soc 1991; 39:1105–1109. 8. Mylotte JM, Karuza J, Bentley DW. Methicillin-resistant Staphylococcus aureus: A questionnaire survey of 75 long-term care facilities in western New York. Infect Con- trol Hosp Epidemiol 1992; 13:711–718. 9. Ward TT, Strausbaugh LJ. Increasing prevalence of methicillin-resistant Staphylo- coccus aureus in hospitals and nursing homes: The Oregon experience. Infect Med 1992; 9:46–51. 10. Strausbaugh LJ, Crossley KB, Nurse BA, Thrupp LD, SHEA Long-Term-Care Com- mittee. Antimicrobial resistance in long-term-care facilities. Infect Control Hosp Epi- demiol 1996; 17:129–140.

MRSA 405 11. Storch GA, Radcliff JL, Meyer PL, Hinrichs JH. Methicillin-resistant Staphylococcus aureus in a nursing home. Infect Control 1987; 8:24–29. 12. Thomas JC, Bridge J, Waterman S, Vogt J, Kilman L, Hancock G. Transmission and control of methicillin-resistant Staphylococcus aureus in a skilled nursing facility. In- fect Control Hosp Epidemiol 1989; 10:106–110. 13. Muder RR, Brennen C, Wagener MM, Vickers RM, Rihs JD, Hancock GA, Yee YC, Miller JM, Yu VL. Methicillin-resistant staphylococcal colonization and infection in a long-term care facility. Ann Intern Med 1991; 114:107–112. 14. Strausbaugh LJ, Jacobson C, Sewell DL, Potter S, Ward TT. Methicillin-resistant Staphylococcus aureus in extended-care facilities: Experiences in a Veterans’ Affairs nursing home and a review of the literature. Infect Control Hosp Epidemiol 1991; 12:36–45. 15. Hsu CCS. Serial survey of methicillin-resistant Staphylococcus aureus nasal carriage among residents in a nursing home. Infect Control Hosp Epidemiol 1991; 12:416– 421. 16. Bradley SF, Terpenning MS, Ramsey MA, Zarins LT, Jorgensen KA, Sottile WS, Schaberg DR, Kauffman CA. Methicillin-resistant Staphylococcus aureus: Colo- nization and infection in a long-term care facility. Ann Intern Med 1991; 115: 417–422. 17. Murphy S, Denman S, Bennett RG, Greenough WB III, Lindsay J, Zelesnick LB. Me- thicillin-resistant Staphylococcus aureus colonization in a long-term-care facility. J Am Geriatr Soc 1992; 40:213–217. 18. Terpenning MS, Bradley SF, Wan JY, Chenoweth CE, Jorgensen KA, Kauffman CA. Colonization and infection with antibiotic-resistant bacteria in a long-term care facil- ity. J Am Geriatr Soc 1994; 42:1062–1069. 19. Mulhausen PL, Harrell LJ, Weinberger M, Kochersberger GG, Feussner JR. Con- trasting methicillin-resistant Staphylococcus aureus colonization in Veterans Affairs and community nursing homes. Am J Med 1996; 100:24–31. 20. Lee Y-L, Gupta G, Cesario T, Lee R, Nothvogel S, Nassar J, Flionis L, Thrupp L. Colonization by Staphylococcus aureus resistant to methicillin and ciprofloxacin dur- ing 20 months’ surveillance in a private skilled nursing facility. Infect Control Hosp Epidemiol 1996; 17:649–653. 21. Lee Y-L, Cesario T, Gupta G, Filonis L, Tran C, Decker M, Thrupp L. Surveillance of colonization and infection with Staphylococcus aureus susceptible or resistant to methicillin in a community skilled-nursing facility. AJIC Am J Infect Control 1997; 25:312–321. 22. Hsu CCS, Macaluso CP, Special L, Hubble RH. High rate of methicillin-resistance of Staphylococcus aureus isolated from hospitalized nursing home patients. Arch Intern Med 1988; 148:569–570. 23. Budnick LD, Schaefler S, New York MRSA Study Group. Ciprofloxacin-resistant methicillin-resistant Staphylococcus aureus in New York health care facilities, 1988. J Public Health 1990; 80:810–813. 24. Flournoy DJ. Antimicrobial susceptibilities of bacteria from nursing home residents in Oklahoma. Gerontology 1994; 40:53–56. 25. Smith PW, Selp CW, Schaefer SC, Bell-Dixon C. Microbiologic survey of long-term care facilities. AJIC Am J Infect Control 2000; 28:8–13.

406 Strausbaugh 26. Stover BH, Duff A, Adams G, Buck G, Hancock G, Rabalais G. Emergence and con- trol of methicillin-resistant Staphylococcus aureus in a children’s hospital and pedi- atric long-term care facility. AJIC Am J Infect Control 1992; 20:248–255. 27. Preston M, Borczyk A, Jamieson F. Epidemic methicillin-resistant Staphylococcus aureus strain—Ontario. Can Commun Dis Rep 1998; 15:47–49. 28. Morrison L, Stolarek I. Does MRSA affect patient outcomes in the elderly? A retro- spective pilot study. J Hosp Infect 2000; 45:169–171. 29. Aeilts GD, Sapico FL, Canawati HN, Malik GM, Montgomerie JZ. Methicillin-resis- tant Staphylococcus aureus colonization and infection in a rehabilitation facility. J Clin Microbiol 1982; 16:218–223. 30. Strausbaugh LJ, Jacobson C, Yost T. Methicillin-resistant Staphylococcus aureus in a nursing home and affiliated hospital: A four-year perspective. Infect Control Hosp Epidemiol 1993; 14:331–336. 31. Boyce JM. Methicillin-resistant Staphylococcus aureus in hospitals, long-term care facilities: Microbiology, epidemiology, and preventive measures. Infect Control Hosp Epidemiol 1992; 13:725–737. 32. Mulligan ME, Murray-Leisure KA, Ribner BS, Standiford HC, John JF, Korvick JA, Kauffman CA, Yu VL. Methicillin-resistant Staphylococcus aureus: A consensus re- view of the microbiology, pathogenesis, and epidemiology with implications for pre- vention and management. Am J Med 1993; 94:313–328. 33. Boyce JM, Jackson MM, Pugliese G, Batt MD, Fleming D, Garner JS, Hartstein AI, Kauffman CA, Simmons M, Weinstein R, Williams CO, AHA Technical Panel on In- fections within Hospitals. Infect Control Hosp Epidemiol 1994; 15:105–115. 34. Bradley SF. Issues in the management of resistant bacteria in long-term-care facili- ties. Infect Control Hosp Epidemiol 1999; 20:363–366. 35. Larsen E, Bobo L, Bennett R, Murphy S, Seng ST, Hoo JTE, Sisler J. Lack of care- giver hand contamination with endemic bacterial pathogens in a nursing home. AJIC Am J Infect Control 1991; 19:11–15. 36. Lee YI, Cesario T, Lee R, Nothvogel S, Nassar J, Farsad N, Thrupp L. Colonization by Staphylococcus species resistant to methicillin or quinolone on hand of medical personnel in a skilled-nursing facility. AJIC Am J Infect Control 1994; 22:346–351. 37. Strausbaugh LJ, Jacobson C, Sewell DL, Potter S, Ward TT. Antimicrobial therapy for methicillin-resistant Staphylococcus aureus colonization in residents and staff of a Veterans Affairs nursing home care unit. Infect Control Hosp Epidemiol 1992; 13: 151–159. 38. Spindel SJ, Strausbaugh LJ, Jacobson C. Infection caused by Staphylococcus aureus in a Veterans’ Affairs nursing home care unit: A 5-year experience. Infect Control Hosp Epidemiol 1995; 16:217–223. 39. Strausbaugh LJ, Joseph C. Epidemiology and prevention of infections in residents of long-term care facilities. In: Mayhall CG (ed). Hospital Epidemiology and Infection Control, 2nd ed. Philadelphia, Lippincott Williams and Wilkins, 1999:1461–1482. 40. Boyce JM. Methicillin-resistant Staphylococcus aureus in nursing homes: Putting the problem in perspective. Infect Control Hosp Epidemiol 1991; 12:413–415. 41. Brennen C, Muder RR. Conjunctivitis associated with methicillin-resistant Staphylo- coccus aureus in a long-term care facility. Am J Med 1990; 88:5-14N–5-17N.

MRSA 407 42. Bentley DW, Bradley S, High K, Schoenbaum S, Taler G, Yoshikawa TT. Practice guideline for evaluation of fever and infection in long-term care facilities. Clin Infect Dis 2000; 31:640–653. 43. Kloos WE, Bannerman TL. Staphylococcus and Micrococcus. In: Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH (eds). Manual of Clinical Microbiology, 7th ed. Washington, D.C., ASM Press, 1999:264–282. 44. Hubert SK, Mohammed JM, Fridkin SK, Gaynes RP, McGowan JE Jr, Tenover FC. Glycopeptide-intermediate Staphylococcus aureus: Evaluation of a novel screening method and results of a survey of selected U.S. hospitals. J Clin Microbiol 1999; 37: 3590–3593. 45. Fekety R. Vancomycin, teicoplanin, and the streptogramins: Quinupristin and dalfo- pristin. In: Mandell GL, Bennett JE, Dolin R (eds). Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 5th ed. Philadelphia, W. B. Saunders Company, 2000:382–392. 46. Moellering RC Jr. A novel antimicrobial agent joins the battle against resistant bac- teria. Ann Intern Med 1999; 130:155–157. 47. Donowitz GR. Oxazolidinones. In: Mandell GL, Bennett JE, Dolin R (eds). Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 5th ed. Philadelphia, W. B. Saunders Company, 2000:392–394. 48. Chien JW, Kucia ML, Salata RA. Use of linezolid, an oxazolidinone, in the treatment of multidrug-resistant gram-positive bacterial infections. Clin Infect Dis 2000; 30: 146–151. 49. Rubinstein E, Cammarata SK, Oliphant TH, Wunderlink RG, Linezolid Nosocomial Pneumonia Study Group. Linezolid (PNU-100766) versus vancomycin in the treat- ment of hospitalized patients with nosocomial pneumonia: A randomized, double- blind, multicenter study. Clin Infect Dis 2001; 32:402–412. 50. Myers JP, Linneman CC Jr. Bacteremia due to methicillin-resistant Staphylococcus aureus. J Infect Dis 1982; 145:532–536. 51. Muder RR, Boldin M, Brennen C, Hsieh M, Vickers RM, Mitchum K, Yee YC. A controlled trial of rifampicin, minocycline, and rifampicin plus minocycline for erad- ication of methicillin-resistant Staphylococcus aureus in long-term care patients. J Antimicrob Chemother 1994; 34:188–190. 52. Tunkel AR. Topical antibacterials. In: Mandell GL, Bennett JE, Dolin R (eds). Man- dell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 5th ed. Philadelphia, W. B. Saunders Company, 2000:428–435. 53. Cederna JE, Terpenning MS, Ensberg M, Bradley SF, Kauffman CA. Staphylococcus aureus nasal colonization in a nursing home: Eradication with mupirocin. Infect Con- trol Hosp Epidemiol 1990; 11:13–16. 54. Kauffman CA, Terpenning MS, He X, Sarins LT, Ramsey MA, Jorgensen KA, Sot- tile WS, Bradley SF. Attempts to eradicate methicillin-resistant Staphylococcus au- reus from a long-term-care facility with the use of mupirocin ointment. Am J Med 1993; 94:371–378. 55. Watanakunakorn C, Axelson C, Bota B, Stahl C. Mupirocin ointment with and with- out chlorhexidine baths in the eradication of Staphylococcus aureus nasal carriage in nursing home residents. AJIC Am J Infect Control 1995; 23:306–309.

408 Strausbaugh 56. Bennett ME, Thurn JR, Klicker R, Williams CO, Weller M. Recommendations from a Minnesota task force for the management of persons with methicillin-resistant Staphylococcus aureus. AJIC Am J Infect Control 1992; 20:42–48. 57. Bradley SF. MRSA in long-term care: Fact, fiction and controversy. Infect Dis Clin Pract 1994; 3:321–326. 58. Cahill CK, Rosenberg J. Guideline for prevention and control of antibiotic-resistant microorganisms in California long-term care facilities. J Gerontol Nurs 1995; May: 40–47. 59. Yoshikawa TT. VRE, MRSA, PRP, and DRGNB in LTCF: Lessons to be learned from this alphabet. J Am Geriatr Soc 1998; 46:241–243. 60. Bonomo RA, Rice LB. Emerging issues in antibiotic resistant infections in long-term care facilities. J Gerontol A Biol Sci Med Sci 1999; 54A:B260–B267. 61. Sioux Falls Task Force on Antimicrobial Resistance. Guidelines for the prevention and control of methicillin-resistant Staphylococcus aureus in long-term care facili- ties. South Dakota Med J 1999; (52):235–240. 62. Bradley SF. Methicillin-resistant Staphylococcus aureus: Long-term care concerns. Am J Med 1999; 106(5A):2S–10S. 63. Bonomo RA. Multiple antibiotic-resistant bacteria in long-term-care facilities: An emerging problem in the practice of infectious diseases. Clin Infect Dis 2000; 31:1414–1422. 64. Pennington GR. Multiple antibiotic resistant Staphylococcus aureus in rehabilitation and extended care facilities: A non-problem. Aust NZ J Surg 1982; 52:534–535. 65. Bryce EA, Tiffin SM, Isaac-Renton JL, Wright CJ. Evidence of delays in transferring patients with methicillin-resistant Staphylococcus aureus or vancomycin-resistant Enterococcus to long-term-care facilities. Infect Control Hosp Epidemiol 2000; 21: 270–271. 66. Loeb M, Moss L, Stiller A, Smith S, Russo R, Molloy DW, Wodchis W. Coloniza- tion with multiresistant bacteria and quality of life in residents of long-term-care fa- cilities. Infect Control Hosp Epidemiol 2001; 22:67–68. 67. Nicolle LE, Bentley DW, Garibaldi R, Neuhaus EG, Smith PW, SHEA Long-Term- Care Committee. Antimicrobial use in long-term-care facilities. Infect Control Hosp Epidemiol 2000; 21:537–545. 68. Loeb M. Antibiotic use in long-term-care facilities: Many unanswered questions. In- fect Control Hosp Epidemiol 2000; 21:680–683. 69. Simor AE, Augustin A, Ng J, Betschel S, McArthur M. Control of MRSA in a long- term care facility. Infect Control Hosp Epidemiol 1994; 15:69–70. 70. Smith P, Rusnak P. SHEA/APIC Position Paper—infection prevention and control in the long-term-care facility. Infect Control Hosp Epidemiol 1997; 18:831–849. 71. Loeb M, Bentley DW, Bradley S, Crossley K, Garibaldi R, Gantz N, McGeer A, Muder RR, Mylotte J, Nicolle LE, Nurse B, Paton S, Simor AW, Smith P, Straus- baugh L. Development of minimum criteria for the initiation of antibiotics in resi- dents of long-term-care facilities: Results of a consensus conference. Infect Control Hosp Epidemiol 2001; 22:120–124. 72. Garner JS, Hospital Infection Control Practices Advisory Committee. Guideline for Isolation Precautions in Hospitals. Infect Control Hosp Epidemiol 1996; 17:53–80.

MRSA 409 73. Guilhermetti M, Hernandes SED, Fukushigue Y, Garcia LB, Cardoso CL. Effective- ness of hand-cleansing agents for removing methicillin-resistant Staphylococcus au- reus from contaminated hands. Infect Control Hosp Epidemiol 2001; 22:105–108. 74. Lee Y-L, Thrupp L. Genotyping by restriction endonuclease analysis compared to phenotyping by antibiogram for typing methicillin-resistant Staphylococcus aureus strains colonizing patients in a nursing home. Infect Control Hosp Epidemiol 2000; 21:218–221. 75. von Eiff C, Becker K, Machka K, Stammer H, Peters G. Nasal carriage as a source of Staphylococcus aureus bacteremia. N Engl J Med 2001; 344:11–16. 76. Archer GL, Climo MW. Staphylococcus aureus bacteremia—consider the source. N Engl J Med 2001; 344:55–56. 77. Kernodle DS, Kaiser AB. Postoperative infections and antimicrobial prophylaxis. In: Mandell GL, Bennett JE, Dolin R. eds. Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, 5th ed. Philadelphia, W. B. Saunders Company, 2000:3177–3191. 78. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR, Hospital Infection Con- trol Practices Advisory Committee. Guideline for prevention of surgical site infec- tion, 1999. Infect Control Hosp Epidemiol 1999; 20:247–278.



23 Vancomycin (Glycopeptide)- Resistant Enterococci Lona Mody University of Michigan, and Veterans Affairs Ann Arbor Healthcare System, Ann Arbor, Michigan Shelly A. McNeil Dalhousie University, Halifax, Nova Scotia, Canada Suzanne F. Bradley University of Michigan, and Veterans Affairs Ann Arbor Healthcare System, Ann Arbor, Michigan I. EPIDEMIOLOGY AND CLINICAL RELEVANCE A. The Enterococcus: An Overview The Enterococcus is a normal component of the endogenous gastrointestinal and perineal flora. Overall, the ability of enterococci to cause disease (virulence) is limited relative to other common pathogens such as Staphylococcus aureus, group A beta-hemolytic streptococci, or aerobic gram-negative bacilli. As a result, ente- rococcal infections occur primarily when normal host defenses are impaired (1–5). When the host is compromised, the Enterococcus becomes a significant op- portunistic pathogen causing many of the major clinical infectious syndromes af- fecting man. Enterococcus faecalis, and less often E. faecium, are frequent causes of urinary tract infection (UTI), intra-abdominal and pelvic infection, soft tissue infection, bacteremia, and endocarditis. Enterococci commonly coexist with other pathogens in the setting of gastrointestinal and soft-tissue infections. Other ente- rococcal species, E. gallinarum, E. casseliflavus, and others, rarely cause infec- tion (1–4), (Table 1). The emergence of resistance to glycopeptide antibiotics 411

412 Mody et al. Table 1 Prevalence of Enterococcal Species Among Clinical Isolates Glycopeptide-resistant strains (%) All clinical isolates (64) Percent United States (78) Europe (79) Enterococcus faecalis 80–90 2.4 Ͻ 0.1 Enterococcus faecium 5–15 46.9 3.8 Enterococcus casseliflavus Ͻ5 17.6 Enterococcus gallinarum 19.1 Enterococcus durans Enterococcus avium Enterococcus raffinosus Other enterococcal species Source: References 64, 78, 79. (vancomycin and teicoplanin) in enterococci is important primarily because ef- fective treatment for serious infection is so difficult (6–8). B. Significance of Glycopeptide Resistance in Enterococci In the United States, enterococci resistant to glycopeptide antibiotics are found predominantly among seriously ill patients in the acute care setting, primarily in intensive care units, in association with prolonged hospital stays, prolonged use of broad-spectrum antibiotics, and frequent use of invasive devices (9,10). Approx- imately 10% of enterococcal bloodstream isolates from hospitals have been found to be resistant to vancomycin (11). Mortality caused by vancomycin-resistant en- terococci (VRE), particularly bacteremia, has been high. However, it has not been possible to establish vancomycin resistance as an independent risk factor for mor- tality in hospitalized patients (12,13). Increased mortality from VRE has been thought to be the result of the lack of effective antibiotic treatments for infection and to the severity of illness in populations at risk rather than an increase in the virulence of VRE (14). Enterococcus faecalis has been the most common enterococcal species causing 80% to 90% of infections; however, E. faecium is the predominant species manifesting vancomycin resistance (2,4) (Table 1). In the United States, VRE are found infrequently in non-intensive care unit and outpatient settings (15,16). Healthy healthcare workers have rarely been found to be colonized with van- comycin-resistant E. faecium or E. faecalis (9). In parts of Europe, VRE com- monly colonizes healthy humans and pets in the community, but most are rarely pathogenic enterococcal species (17,18). It is thought that VRE have emerged because of the selective pressure of an- tibiotic use. Enterococci resistant to antibiotics, such as glycopeptides, exist in na-

Vancomycin-Resistant Enterococci 413 ture, albeit in small numbers. These clones are selected in the gastrointestinal tract when a patient is exposed to antibiotics and normal flora is suppressed, allowing resistant enterococci to emerge. In Europe, community-based strains emerged ini- tially in livestock because of the use of glycopeptide antibiotics, such as avoparicin, in animal feeds (6). Ingestion of meat contaminated with these ente- rococci may have contributed to widespread colonization in humans. In the United States, specific antibiotics, such as the glycopeptides themselves, third-generation cephalosporins, and antibiotics with anaerobic activity, have been particularly as- sociated with the emergence of VRE (19). Enterococci also thrive readily on inan- imate surfaces. In hospital, patients may acquire antibiotic-resistant enterococci from other VRE-colonized patients or from contaminated hands of healthcare workers or environmental sources (9,10). C. Resistant Enterococci in Nursing Homes: Not a New Problem Residents of long-term care facilities (LTCFs) are frequently found to harbor an- tibiotic-resistant bacteria upon admission to hospital and may represent a reservoir for these organisms (20). For instance, outbreaks in hospitals have been frequently traced to residents of LTCFs who were found to be colonized upon admission to acute care (21–23). Frequent colonization with antibiotic-resistant enterococci in LTCF residents is not a new problem. Studies of high-level gentamicin-resistant enterococci (HGRE) done a decade earlier in long-term care were predictive of many of the risk factors for VRE later identified in that setting. High-level gen- tamicin-resistant enterococci colonization was common in residents of LTCFs, with rates of 35% to 47% observed in a single nursing home (24,25). More resi- dents were already colonized with HGRE at the time of admission (~22%) than those who acquired it (14%) during their stay in the nursing home (25). Many of the LTCF strains were closely related to strains found in the attached acute care facility, suggesting that acquisition might have occurred in hospital (24). In one LTCF, residents colonized with HGRE were more likely to have poor functional status, wounds, prior antibiotic therapy, colonization with methicillin- resistant Staphylococcus aureus (MRSA), and require urethral catheterization than uncolonized residents (24–26). Rectum, wounds, and perineum were the most common sites of colonization (10,11). Over a 3-year period, HGRE ac- counted for 6% to 11% of infections occurring in 4% of residents, most of whom had previously been colonized (10,11). Most of those infections were urinary tract or soft tissue infections that were not severe. D. The Epidemiology of VRE in LTCFs: What is Known The prevalence of VRE carriage in residents of LTCFs is only now being estab- lished. Most of the information on VRE in the LTCF is based on studies of asymp-

414 Mody et al. tomatic rectal carriage rather than infection. In the acute care setting, the preva- lence of VRE infection has varied widely with geographic area. Rates of colo- nization with VRE in LTCFs are likely to parallel those seen in local hospitals. In three studies of nursing home residents admitted to hospital, rates of colonization varied from 10% to 47% (21–23). Surveillance of patients in an acute care hospi- tal without a VRE infection problem revealed 18 carriers of vancomycin-resistant E. faecium and three carriers of E. gallinarum on acute care wards and only one carrier of E. faecalis on the chronic care ward (27). Point prevalence surveys for VRE colonization in residents of a Michigan LTCF were performed over 2 years at 6-month intervals. The prevalence of VRE colonization in rectum ranged from 9% to 22% with infrequent colonization of wounds (28). Residents of LTCFs colonized with VRE have significant debility. Many have significant comorbid illnesses or wounds (33%), urinary devices (47%), or feeding tubes (22%), and were frequently co-colonized with MRSA (47%) or Clostridium difficile (19%) (29). At least half the colonized patients had received recent treatment with vancomycin, a cephalosporin, or both (29). Many of the residents colonized with VRE had been recently hospitalized and may have introduced VRE into the LTCF (28–33). In one study, 67% of residents were al- ready positive for VRE upon admission to the LTCF (29). Residents of LTCFs with VRE were four times more likely to have been recently discharged from hos- pitals where the organism was endemic than uncolonized residents (30). Strains obtained from those LTCF residents were closely related genetically to VRE strains that predominated in the transferring hospitals (30). All VRE identified in an LTCF may not represent the same organism or be proof that spread is occurring. Multiple, rather than single, strains often circulate in an LTCF at the same time (28,34). In addition, individual LTCF residents have been shown to carry multiple strains of VRE at the same time, which then emerge when antibiotic therapy is initiated (34). The prevalence of VRE may remain high in LTCFs because carriage can persist for months (28,29) and can be prolonged by the use of antimicrobial therapy (29). Roommates had not been shown to share the same strain of VRE in one study, but spread could occur from the environment or the hands of personnel who were frequently colonized with multiple strains of VRE (28,32). Data on VRE infection, rather than colonization in LTCF residents, are scant. A 1992 surveillance study of clinical isolates submitted to an Oklahoma laboratory from 100 nursing homes revealed that 3% of 243 E. faecalis and 12% of 32 E. fae- cium were resistant to vancomycin (35). Whether these clinical isolates represented true infection or asymptomatic carriage could not be determined by the design of the study. In prospective surveillance studies, severe infection with VRE has been uncommon, relative to rates of colonization. Over 2 years or more of surveillance, seven UTIs, one bacteremia, and no deaths attributable to VRE infection were noted in two nursing homes where VRE colonization was common (28,29).

Vancomycin-Resistant Enterococci 415 II. CLINICAL MANIFESTATIONS Enterococci have the potential to cause infection in older adults residing in LTCFs; however, the prevalence of VRE infection is unknown. The clinical man- ifestations of vancomycin-susceptible enterococci and VRE are identical (Table 2). Urinary tract infection (UTI) is the most common infection in LTCF residents (36). Gram-negative bacilli clearly predominate as the major causes of UTI in res- idents of LTCFs, but infections with enterococci are more common in older adults requiring hospitalization than in healthy community dwellers (37). Enterococci have been reported to cause 2% to 13% of UTIs in the long-term care setting (38–43). Urethral catheterization may be an important risk factor for enterococcal UTIs in LTCF residents as almost 20% of this population had enterococci colo- nizing or infecting their urine (44). Skin and soft tissue infections associated with enterococci are rarely men- tioned in the long-term care setting (39,41). Despite the facts that enterococci have been isolated from 7% to 46% of diabetic polymicrobial foot infections and that diabetes mellitus and its complications are common in older adults, enterococcal infections have been rarely reported in series of soft tissue infection among resi- dents of LTCFs (39,41,45). Enterococci have also been recovered from the polymicrobial flora of pressure ulcers (36). Whether enterococci are significant pathogens in polymicrobial soft tissue infections and require specific antimicro- bial therapy is controversial (45–48). Aging is associated with increased prevalence of hepatobiliary disease, di- verticulosis, and other gastrointestinal pathology with a risk of infectious compli- cations (49,50). Polymicrobial infection, including enterococci from biliary sources and after intra-abdominal surgery, is not uncommon (49–51). Enterococ- cal infections specifically associated with intra-abdominal/gastrointestinal sources have not been noted in surveys of infection in LTCFs; however, clinicians should be aware that older adults may develop intra-abdominal sources of entero- coccal infection during their stay. Table 2 Clinical Syndromes Associated with Enterococci Syndrome Asymptomatic colonization Urinary tract infection Skin/soft tissue infection Intra-abdominal infection Bloodstream infection Endocarditis Meningitis/pneumonia (true infections extremely rare)

416 Mody et al. Overall, bacteremia accounts for 2.7% to 16.3% of nursing home-acquired infections in different series (39,41,52,53). Reports of the prevalence of bac- teremia in LTCFs may be misleading, as many facilities may not have the re- sources to obtain routine blood cultures. Most bloodstream infections originated from a urinary source (55% to 56%) or less often from a soft tissue infection (7%–14%); gastrointestinal sources were rarely described (52,54,55). Gram-pos- itive cocci have accounted for 33% to 35% of bloodstream isolates, with S. aureus predominating in most series (36,52,54,55). Enterococci were noted in 3.3% to 9.1% of bloodstream infections (31–34). Isolation of enterococci occurred most commonly in the setting of polymicrobial bacteremia with another organism. En- terococcal bacteremia from a urinary source was often associated with the use of a urinary device (52,56,57). Because bacteremia appears to be an uncommon event in LTCFs, it might be assumed that metastatic seeding of a distant site with enterococci, such as a heart valve, would be unlikely. However, the instrumentation of a colonized uri- nary tract or gastrointestinal tract, typically in an older man, with subsequent transient bacteremia and seeding of a native heart valve is the classic scenario for the development of enterococcal endocarditis. Infection with enterococci has been reported in 7% to 20% of older adults with endocarditis (58–62). Whether enterococcal endocarditis occurs more often in older adults as a consequence of increased frequency of genitourinary and gastrointestinal pathology, instrumen- tation, or predisposing valvular disease remains a subject of debate (58–62). III. DIAGNOSTIC APPROACH Enterococci can be easily isolated from cultures of urine, stool, wounds, blood, ab- scess material, and rectal swabs using standard culture media. It is important for the laboratory to speciate all enterococci and screen for the presence of vancomycin resistance. Detection of vancomycin resistance in colonizing or infecting strains of enterococci requires the use of appropriate microbiological methods. Standard broth microdilution methods, disk diffusion, or E-test methods can be used. How- ever, some automated methods are unreliable in detecting VRE (63,64). Different VRE species also vary in their antimicrobial susceptibility patterns (Tables 1,3,4). Patterns of resistance to vancomycin and teicoplanin and the level of resis- tance to those antibiotics have been used as phenotypic markers for the five mech- anisms of resistance currently described in enterococci (Table 3). These five VRE phenotypes are termed VanA, VanB, VanC, VanD, and VanE. These phenotypes provide additional information regarding the likelihood that vancomycin resis- tance might spread or respond to certain antibiotics (64–66). Vancomycin resistance in enterococci is defined by a minimum inhibitory concentration (MIC) of 2 ␮g/ml or more. Resistance to vancomycin at low levels (MIC, 2–32 ␮g/ml) and susceptibility to teicoplanin is typical of E. casseliflavus,

Vancomycin-Resistant Enterococci 417 Table 3 Phenotypes of Vancomycin-Resistant Enterococci Based on Antimicrobial Susceptibilities to Vancomycin and Teicoplanin Phenotype Resistance Common Vancomycin- Teicoplanin- element species resistance susceptible VanA Acquired/transferable Enterococcus High level No faecium, E. VanB Acquired/transferable faecalis High level Yes VanC Low level Yes VanD Intrinsic/not E. faecium, E. High level No VanE transferable faecalis Low level Yes Acquired/not E. gallinarum, transferable E. casseliflavus E. faecium Acquired/not transferable E. faecalis E. gallinarum, and E. flavescens. These species are referred to phenotypically as VanC strains, and rarely cause clinically significant disease. VanC-mediated re- sistance is an intrinsic and chromosomally mediated characteristic of these organ- isms that is not transferable to other bacteria (64–66). Vancomycin resistance also can be acquired from other organisms by some enterococci. VanA and VanB strains are found most commonly and are important epidemiologically because they can spread or transfer vancomycin resistance ele- ments to other bacteria. Acquisition of resistance elements can lead to high-level resistance to vancomycin (MIC, Ն64 ␮g/ml) and teicoplanin (MIC, Ն16 ␮g/ml) in strains that have required the VanA gene. Acquired resistance to vancomycin with susceptibility to teicoplanin is referred to as a VanB strain. VanA and VanB strains are most often found among strains of E. faecium and E. faecalis. VanA strains are found widely throughout the United States, with VanB strains present on a regional basis (64–66). The diagnosis of infection with VRE is based on the isolation of the organ- ism in association with symptoms and signs consistent with an appropriate clini- cal syndrome (2,5) (Table 2). The clinical presentation of these syndromes is ad- dressed elsewhere. Isolation of VRE in the absence of clinically apparent symptoms or signs represents asymptomatic colonization of urine, skin, or stool, or contamination of wounds or blood cultures. In the appropriate clinical setting, vancomycin-resistant E. faecium and E. faecalis are more likely to represent true pathogens (7). Isolation of E. casseliflavus, E. gallinarum, E. flavescens, and other species likely represents colonization, unless obtained from a sterile site, on mul- tiple occasions, and in high inoculum (5,7,64).

418 Mody et al. Table 4 Treatment of Vancomycin-Resistant Enterococcal Infection* Antibiotic Route Indication Ampicillin IV, PO Efficacious in susceptible Enterococcus faecalis strains Quinupristin/Dalfopristin IV Bactericidal for E. faecalis in combination Linezolid PO, IV with gentamicin or streptomycin unless Nitrofurantoin PO high-level aminoglycoside resistance is Doxycycline PO, IV present Quinolones PO, IV Chloramphenicol IV E. faecium strains generally resistant to Teicoplanin IV normal regimens of ampicillin High-dose IV ampicillin/beta-lactamase inhibitor regimens experimental E. faecium, only use in serious infections E. faecalis not susceptible Bacteriostatic agent for enterococci Toxicities common: myalgias, phlebitis Resistance described, but rare E. faecium or E. faecalis, only use in serious infections Oral formulation 100% bioavailable. Bacteriostatic Efficacious for urinary tract infection only May be effective in treatment of urinary tract infections. Efficacy in serious infections unpredictable May be effective in treatment of urinary tract infections Efficacy in serious infections unpredictable Many E. faecium susceptible Efficacy in serious infection not established Significant hematologic toxicities Serious infections VanB strains only Not available in the United States * Enterococcal strains must demonstrate sensitivity to an agent using approved antimicrobial susceptibility methods. Abbreviations: IV, intravenous; PO, oral (per os). IV. THERAPEUTIC INTERVENTIONS Compared with other gram-positive cocci, the Enterococcus is relatively resistant to the bactericidal effects of cell wall-active antibiotics. Even among vancomycin- susceptible enterococci, intrinsic resistance to many antibiotic classes is common. Penicillins and vancomycin remain the most reliable treatments for infections

Vancomycin-Resistant Enterococci 419 caused by susceptible enterococci, but their activities are bacteriostatic rather than bactericidal. Only the addition of an aminoglycoside to vancomycin or penicillins provides reliable and effective bactericidal activity for the treatment of serious en- terococcal infections (6–8,64–68). Unfortunately, resistance to vancomycin in enterococci is usually associ- ated with resistance to multiple antibiotics, including penicillins and aminoglyco- sides. Most vancomycin-resistant E. faecium and many E. faecalis are resistant to normally achievable levels of penicillin and ampicillin and high-levels of gen- tamicin or streptomycin (6–8,64–68) (Table 4). In the event of resistance to peni- cillin in VRE, antimicrobial susceptibilities to other antibiotic classes should be assessed. If susceptible, nitrofurantoin may be effective in treating a UTI caused by VRE. However, despite in vitro susceptibility to tetracyclines, chlorampheni- col, and quinolones, clinical success in the treatment of serious VRE infections has been infrequent (6–8,64–68). Newer agents may be effective in treating milder infections, but their an- timicrobial activity remains bacteriostatic. A new streptogramin, quinpristin/dal- fopristin (Synercid®) is active against E. faecium but not E. faecalis, whereas the oxazolidinone, linezolid (Zyvox®), is active against both species. Teicoplanin is active only against VanB and VanC strains but is not approved for use in the United States. Newer antibiotic classes active against VRE, such as the lipogly- copeptides (ramoplanin), the acidic lipopeptides (daptomycin), and glycylcy- clines are under investigation. Many of the agents have significant toxicities and may not be bactericidal for the Enterococcus. Surgical incision and drainage with removal of foreign devices, whenever possible, remains a mainstay of treatment for infections caused by VRE (6–8,64–68). V. INFECTION CONTROL MEASURES In the acute care setting, VRE infections lead to increased morbidity and increased costs of treatment. The extensive use of infection control resources in the hospital setting can, therefore, be easily justified. It is not clear that VRE is a cause of se- rious infection or that transmission of VRE is common in the LTCF setting. In ad- dition, a significant proportion of residents of LTCFs may be colonized with more than one drug-resistant bacterium for prolonged periods. Hospital-based infection control procedures, such as long-term confinement in a private room, if imple- mented in an LTCF, would have a significant impact on the psychological, social, and physical needs of the residents. The controversies surrounding the control of VRE and other antibiotic-re- sistant bacteria in LTCFs have been addressed by the Society for Healthcare Epi- demiology of America (SHEA) Committee on Long-Term Care (69,70) (Table 5). Their infection control recommendations are modifications of Contact Precau-

420 Mody et al. Table 5 Strategies and Procedures for the Control of VRE in LTCFs Strategies/Procedures Employee education Surveillance of cultures obtained for clinical reasons/symptomatic infection Establishes the rate of VRE infections in an individual LTCF Establishes what is the normal infection rate for an LTCF Defines when an infection rate is abnormal and potential epidemic transmission Defines when to start procedures to control an outbreak Maintain listing of VRE carriers that are already known Useful information if an outbreak of infection suspected in LTCFs or hospitals Transferring facilities should routinely provide this information to receiving facilities, if known Use of routine surveillance cultures specifically to detect asymptomatic VRE colonization May be falsely reassuring if negative Increased cultures, need for isolation not cost effective unless documented that infections are prevented Isolation procedures Private room/cohorting with other colonized residents recommended, but efficacy in LTCFs not established VRE-colonized residents with good hygiene, no diarrhea or draining wounds may room/share bathrooms with uncolonized residents who are not severely compromised, do not have urinary catheters, drainage devices, or wounds, and are not on broad- spectrum antibiotics VRE-colonized resident with good hygiene, no diarrhea and draining wounds contained by a dressing need not be confined to their rooms Isolation can be discontinued after two successive negative cultures of stool or wounds Hand washing Mandatory before and after caring for all residents Antimicrobial soaps and hand disinfectants suggested, but efficacy not established in LTCFs Gloves/gowns Use if contact with body fluids for all residents Use in the room before contact with a VRE-colonized or infected resident or his inanimate environment Gowns recommended if contamination of healthcare worker clothes likely Environmental disinfection Daily cleaning of room surfaces and equipment recommended, but efficacy and optimum germicide not established in LTCFs Dedicated equipment for VRE-colonized or infected residents, if available Antibiotic use Monitor antibiotic use Reduce unnecessary use of antibiotics

Vancomycin-Resistant Enterococci 421 Table 5 (Continued) Strategies/Procedures Follow guidelines for appropriate use of vancomycin Serious gram-positive infections resistant to beta-lactam antibiotics Serious allergies to beta-lactam antibiotics Clostridium difficile infections unresponsive to metronidazole Prophylaxis for residents at high risk of endocarditis Prophylaxis for surgical procedures with prosthetic devices and risk of methicillin- resistant staphylococcal infection Limit vancomycin prophylaxis to only two doses Decolonization regimens for VRE Frequent relapses No evidence of efficacy, particularly in residents of LTCFs Emergence of resistance likely Abbreviations: VRE, Vancomycin-resistant enterococci; LTCF, Long-term care facility. Source: Ref. 69. tions for drug-resistant bacteria recommended for acute care facilities (71,72) (see Chapter 8). These precautions acknowledge the limited infection control re- sources in nursing homes and those that are achievable in facilities that provide long-term care. A. Screening for VRE The essential element of an infection control program to minimize the spread of VRE includes the routine use of barrier precautions in all residents of LTCFs. Routine screening cannot completely exclude that VRE is present in stool at low levels. Detection of VRE colonization may occur only under certain circum- stances, such as during therapy with specific antibiotic classes (19). The SHEA guidelines recommend discontinuation of VRE precautions in an LTCF if two rec- tal or wound cultures are negative on successive days. However, if VRE are pre- sent in referring hospitals or in the LTCF itself, it may be prudent to assume that all residents are potential VRE carriers. Moreover, given the limits of detection of VRE, it may not be reasonable to accept residents with negative stool cultures for VRE into a nursing home while refusing others with positive cultures if VRE is clearly known to be present in referring institutions. Routine use of screening pro- cedures to detect carriers for the purposes of elimination of VRE from an LTCF in an endemic geographic locale is unlikely to be cost effective and is not recom- mended. Routine surveillance cultures to detect VRE colonization in residents of LTCFs is only recommended if rates of VRE infection are increasing despite rou- tine infection control precautions and transmission is suspected (69).

422 Mody et al. B. Controlling VRE Transmission In light of scant data regarding transmission of VRE in LTCFs, the SHEA guide- lines recommend that VRE-colonized or -infected patients should optimally be placed in a private room or share a room with a roommate colonized with the same organism (69). Given that residents may be colonized with various combinations of VRE, MRSA, resistant gram-negative bacilli, and C. difficile, these isolation recommendations can pose significant logistical problems for infection control professionals in LTCFs. The SHEA guidelines alternatively recommend that VRE-colonized residents can be placed with or share bathrooms with noncolo- nized individuals if the colonized resident is continent of stool and does not have diarrhea or open wounds. In addition, the noncolonized roommate should not be severely compromised; receiving broad-spectrum antibiotics; or have a urinary catheter, drainage device, or open wounds (69). Careful hand washing by colo- nized and noncolonized residents is emphasized. Restriction to rooms is not rec- ommended in residents who are continent, use good hygiene, and have draining wounds contained by bandages. Private rooms or cohorting techniques have been used in LTCFs, but most of these facilities allowed VRE-colonized residents to participate freely in social activities (32,33,73). Direct contact between colonized and noncolonized residents was restricted in only one facility (73). Lack of further transmission of VRE could not be directly attributed to the use of these cohorting techniques. Precautions to disrupt the transmission of antibiotic-resistant pathogens rec- ognize body fluids and the environment as major reservoirs of VRE and the hands of personnel as potential vectors. Therefore, routine hand washing by personnel before and after providing care to a resident is essential. Some studies have shown that standard soaps may not remove VRE from hands as effectively as hand dis- infectants with antimicrobial activity (10). Antimicrobial soaps and alcohol disin- fectants have been used in LTCFs with VRE, but whether these interventions are effective in preventing transmission has not been established (28,32,33,73). It is recommended that clean, nonsterile gloves be worn when contact with body fluids from any patient is likely, as part of standard infection control pre- cautions (71). In an LTCF, it has been recommended that gloves also be worn within the room of a VRE-colonized or infected resident before initiating any di- rect contact with the patient or inanimate environment (69). The effectiveness of infection control by donning gowns before entry into the rooms of VRE-colo- nized or -infected patients in acute care hospitals remains controversial (10). In an LTCF, it has been recommended that gowns be worn only if the clothes of healthcare workers are likely to become soiled with body fluids (69). In uncon- trolled studies, no transmission of VRE was documented in six LTCFs where the use of gowns and gloves was required (29,32,33,73). In one facility, gloves were also required for any casual contact with colonized residents outside of

Vancomycin-Resistant Enterococci 423 their room (73). However, universal gloving for care of uncolonized as well as colonized resident in an LTCF may be just as effective as identifying and re- stricting colonized patients to their rooms and requiring the use of gowns and gloves in preventing transmission of VRE, MRSA, and multidrug-resistant gram-negative bacilli (74). Because VRE is most likely transmitted by environmental sources, the SHEA guidelines recommend the use of dedicated equipment for colonized or in- fected patients (69). The optimum methods and frequency of disinfection have yet to be defined in hospitals or LTCFs, but daily cleaning of environmental surfaces within the resident’s room with an appropriate germicide was recommended (10,69). In uncontrolled studies of four LTCFs, environmental cleaning of the res- ident’s room ranged from thrice weekly to twice daily with a quaternary ammo- nium compound or germicide (32,73). Small equipment/wheelchairs were left in the rooms (32,33) and/or wheelchairs were decontaminated with 1:10 dilution of bleach twice daily (33,73). More randomized, controlled trials are necessary to define the minimum, least costly, and most effective means of infection control in LTCFs. Despite the diversity of infection control measures used above, the superiority of one approach in preventing colonization has not been established (29,32,33,73). No infections or deaths could be attributed directly to VRE in these studies (29,32,33,73). VI. PREVENTION In the hospital setting, oral antimicrobial agents such as bacitracin, doxycycline, and novobiocin alone or in combination, have been used to eradicate VRE from urine, stool, or wounds with frequent recurrences and emergence of further resis- tance (7,8,64). As a result, experts have generally not recommended VRE decol- onization except in patient populations at extreme risk of infection (8). In the LTCF setting, decolonization of VRE-colonized residents with oral bacitracin regimens has been tried with variable success (33,73). However, it should be rec- ognized that many VRE-colonized residents clear their colonization sponta- neously after decolonization failed or without antibiotics (25,28,32,73). It is likely that increased antibiotic use will perpetuate the problem of VRE. Prevention of the emergence of VRE in susceptible populations requires reducing the unnecessary use of antibiotics in animal feed, in hospital, and in the LTCF set- ting. The Hospital Infection Control Practices Advisory Committee (HICPAC) recommends that vancomycin use be limited to treatment of serious gram-positive infections resistant to beta-lactam antibiotics, treatment of patients with serious beta-lactam allergy, and treatment of C. difficile unresponsive to metronidazole (75). Brief courses of vancomycin prophylaxis should be limited to high-risk en-

424 Mody et al. docarditis prevention or surgical procedures involving prosthetic devices with high rates of infection caused by MRSA (6,75). In the hospital setting, restricting the use of third-generation cephalosporins and antibiotics with antianaerobic ac- tivity has been associated with declines in rates of VRE infection (64,68). Further studies need to be done to ascertain if antibiotic restriction leads to declines in the burden of VRE colonization in residents of LTCFs. Antibiotics are frequently continued in the long-term care setting even if cultures are negative, the infection is resistant to the antibiotic chosen, or the organism is likely to represent contam- ination (76). Significant improvements in antibiotic use in the long-term care set- ting can be achieved in the meantime. Modification of risk factors that predispose residents of LTCFs to colonization and infection with antibiotic-resistant bacteria, such as use of feeding tubes, urinary catheters, intravenous devices, presence of wounds, and C. difficile infection, can be made with appropriate geriatric assess- ments and interventions (77). REFERENCES 1. Kaye D. Enterococci: Biologic and epidemiologic characteristics and in vitro suscep- tibility. Arch Intern Med 1982; 142:2006–2009. 2. Murray BE. The life and times of the Enterococcus. Clin Microbiol Rev 1990; 3:46– 65. 3. Murray BE, Weinstock GM. Enterococci: New aspects of an old organism. Proc As- soc Am Physicians 1999; 111:328–334. 4. Moellering RC. Emergence of Enterococcus as a significant pathogen. Clin Infect Dis 1992; 14:1173–1178. 5. Patterson JE, Sweeney AH, Simms M, Carley N, Mangi R, Sabetta J, Lyons RW. An analysis of 110 serious enterococcal infections: Epidemiology, antibiotic susceptibil- ity, and outcome. Medicine 1995; 74:191–200. 6. Murray BE. Vancomycin-resistant enterococci. Am J Med 1997; 101:284–293. 7. Moellering RC. Vancomycin-resistant enterococci. Clin Infect Dis 1998; 26:1196– 1199. 8. Murray BE. Vancomycin-resistant enterococcal infections. N Engl J Med 2000; 342: 710–721. 9. Mayhall CG. The epidemiology and control of VRE: Still struggling to come of age. Infect Control Hosp Epidemiol 1999; 20:650–652. 10. Hayden MK. Insights into the epidemiology and control of infection with van- comycin-resistant enterococci. Clin Infect Dis 2000; 31:1058–1065. 11. Sahm DF, Marsilio MK, Piazza G. Antimicrobial resistance in key bloodstream bac- terial isolates: Electronic surveillance with the Surveillance Network Database— USA. Clin Infect Dis 1999; 29:259–263. 12. Lautenbach E, Bilker WB, Brennan PJ. Enterococcal bacteremia: Risk factors for vancomycin resistance and predictors of mortality. Infect Control Hosp Epidemiol 1999; 20:318–323.

Vancomycin-Resistant Enterococci 425 13. Garbutt JM, Ventrapragada M, Littenberg B, Mundy LM. Association between resis- tance to vancomycin and death in cases of Enterococcus faecium bacteremia. Clin In- fect Dis 2000; 30:466–472. 14. Mundy LM, Sahm DF, Gilmore M. Relationships between enterococcal virulence and antimicrobial resistance. Clin Microbiol Rev 2000; 13:513–522. 15. Archibald L, Phillips L, Monnet D, McGowan JE, Tenover F, Gaynes R. Antimicro- bial resistance in isolates from inpatients and outpatients in the United States: In- creasing importance of the intensive care unit. Clin Infect Dis 1997; 24:211–215. 16. Fridkin SK, Steward CD, Edwards JR, Pryor ER, McGowan JE, Archibald LK, Gaynes RP, Tenover FC. Project Intensive Care Antimicrobial Resistance Epidemi- ology (ICARE) Hospitals. Clin Infect Dis 1999; 29:245–252. 17. Van den Bogaard AE, Stobberingh EE. Epidemiology of resistance to antibiotics: Links between animals and humans. Int J Antimicrob Agents 2000; 14:327–335. 18. Van den Braak N, Ott A, van Velkum A, Kluytmans JAJW, Koeleman JGM, Span- jaard L, Voss A, Weersink AJL, Vandenbroucke-Grauls CMJE, Buiting AGM, Ver- brugh HA, Endtz HP. Prevalence and determinants of fecal colonization with van- comycin-resistant Enterococcus in hospitalized patients in the Netherlands. Infect Control Hosp Epidemiol 2000; 21:5520–5524. 19. Donskey CJ, Chowdhry TK, Hecker MT, Hoyen CK, Hanrahan JA, Hujer AM, Hut- ton-Thomas RA, Whalen CC, Bonomo RA, Rice LB. Effect of antibiotic therapy on the density of vancomycin-resistant enterococci in the stool of colonized patients. N Engl J Med 2000; 343:1925–1932. 20. Bradley SF. Issues in the management of resistant bacteria in long-term-care facili- ties. Infect Control Hosp Epidemiol 1999; 20:362–366. 21. Elizaga ML, Beezhold D, Hayden MK, Rice T, Nathan C, Buccelli C, Ciesielski C. The prevalence of colonization with vancomycin-resistant enterococci among hospi- talized residents of long-term care facilities (abstr). Clin Infect Dis 1996; 23:925. 22. Revuelta MP, Nord JA, Yarrish RL, Kislak JW, Bombardi V. Recycling of hospital- acquired colonization with vancomycin-resistant enterococci via long-term care fa- cilities (abstr). Clin Infect Dis 1995; 21:730. 23. Sargent SJ, Baselski VS, Reed LD, Wong L, Sahm DF. Simultaneous hospital and community emergence of vancomycin-resistant Enterococcus faecium (abstr). Clin Infect Dis 1995; 21:729. 24. Zervos MJ, Terpenning MS, Schaberg DR, Therasse PM, Medendorp SV, Kauffman CA. High-level aminoglycoside-resistant enterococci: Colonization of nursing home and acute care hospital patients. Arch Intern Med 1987; 147:1591–1594. 25. Chenoweth CE, Bradley SF, Terpenning MS, Zarins LT, Ramsey MA, Schaberg DR, Kauffman CA. Colonization and transmission of high-level gentamicin-resistant en- terococci in a long-term care facility. Infect Control Hosp Epidemiol 1994; 15:703– 709. 26. Terpenning MS, Bradley SF, Wan JY, Chenoweth CE, Jorgensen KA, Kauffman CA. Colonization and infection with antibiotic-resistant bacteria in a long-term care facil- ity. J Am Geriatr Soc 1994; 42:1062–1069. 27. Gordts B, Van Landuyt H, Ieven M, Vandamme P, Goossens H. Vancomycin-resis- tant enterococci colonizing the intestinal tracts of hospitalized patients. J Clin Micro- biol 1995; 33:2842–2846.

426 Mody et al. 28. Bonilla HF, Zervos MA, Lyons MJ, Bradley SF, Hedderwick SA, Ramsey MA, Paul LK, Kauffman CA. Colonization with vancomycin-resistant Enterococcus faecium: Comparison of a long-term care unit with an acute-care hospital. Infect Control Hosp Epidemiol 1997; 18:333–339. 29. Brennen C, Wagener MM, Muder RR. Vancomycin-resistant Enterococcus faecium in a long-term care facility. J Am Geriatr Soc 1998; 46:157–160. 30. Trick WE, Kuehnert MJ, Quirk SB, Arduino MJ, Aguero SM, Carson LA, Hill BC, Banerjee SN, Jarvis WR. Regional dissemination of vancomycin-resistant entero- cocci resulting from interfacility transfer of colonized patients. J Infect Dis 1999; 180:391–396. 31. Quale J, Patel K, Zaman M, Saurina G, Hueng W, Odyna V, Ditore D, Landman D. Fecal carriage of vancomycin-resistant enterococci in long-term care facility and hos- pitalized patients. Clin Infect Dis 1995; 21:733–738. 32. Greenaway CA, Miller MA. Lack of transmission of vancomycin-resistant entero- cocci in three long-term-care facilities. Infect Control Hosp Epidemiol 1999; 20:341– 343. 33. Silverblatt FJ, Tibert C, Mikolich D, Blazek-D’ Arezzo J, Alves J, Tack M, Agatiello P. Preventing the spread of vancomycin-resistant enterococci in a long-term care fa- cility. J Am Geriatr Soc 2000; 48:1211–1215. 34. Schoonmaker DJ, Bopp LH, Baltch AL, Smith RP, Rafferty ME, George M. Genetic analysis of multiple vancomycin-resistant Enterococcus isolates obtained serially from two long-term care patients. J Clin Microbiol 1998; 36:2105–2108. 35. Flournoy DJ. Antimicrobial susceptibilities of bacteria from nursing home residents in Oklahoma. Gerontology 1994; 40:53–56. 36. Nicolle LE, Strausbaugh LJ, Garibaldi RA. Infections and antibiotic resistance in nursing homes. Clin Microbiol Rev 1996; 9:1–17. 37. Felmingham D, Wilson APR, Quintana Ai, Gruneberg RN. Enterococcus species in urinary tract infection. Clin Infect Dis 1992; 15:295–301. 38. Sherman FT, Tucci V, Libow LS, Isenberg HD. Nosocomial urinary-tract infections in a skilled nursing facility. J Am Geriatr Soc 1980; 28:456–461. 39. Alvarez S, Shell CG, Woolley TW, Berk SL, Smith JK. Nosocomial infections in long-term facilities. J Gerontol 1988; 43:M9–M17. 40. Magnussen MH, Robb SS. Nosocomial infections in a long-term care facility. Am J Infect Control 1980; 8:12–17. 41. Jacobson C, Strausbaugh LJ. Incidence and impact of infection in a nursing home care unit. Am J Infect Control 1990; 18:151–159. 42. Smith MA, Duke WM. A retrospective review of nosocomial infections in an acute rehabilitative and chronic population at a large skilled nursing facility. J Am Geriatr Soc 1994; 42:45–49. 43. Gambert SR, Duthie EH, Priefer B, Rabinovitch RA. Bacterial infections in a hospi- tal-based skilled nursing facility. J Chron Dis 1982; 35:781–786. 44. Garibaldi RA, Brodine S, Matsumiya S. Infections among patients in nursing homes: Policies, prevalence, and problems. N Engl J Med 1981; 305:731–735. 45. Lipsky BA, Pecoraro RE, Ahroni JH. Foot ulceration and infections in elderly dia- betics. Clin Geriatr Med 1990; 6:747–769.

Vancomycin-Resistant Enterococci 427 46. Temple ME, Nahata MC. Pharmacotherapy of lower limb diabetic ulcers. J Am Geri- atr Soc 2000; 48:822–828. 47. Barie PS, Christou NV, Dellinger P, Rout RW, Stone HH, Waymack JP. Pathogenic- ity of the Enterococcus in surgical infections. Ann Surg 1990; 212:155–159. 48. Nichols RL, Muzik AC. Enterococcal infections in surgical patients: The mystery continues. Clin Infect Dis 1992; 15:72–76. 49. Kahng KU, Roslyn JJ. Surgical issues for the elderly patient with hepatobiliary dis- ease. Surg Clin North Am 1994; 74:345–373. 50. Van Steenbergen W, Rigauts H, Ponette E, Peetermans W, Pelemans W, Fevery J. Percutaneous transhepatic cholecystostomy for acute complicated calculous chole- cystitis in elderly patients. J Am Geriatr Soc 1993; 41:157–162. 51. Olson MM, Allen MO. Nosocomial abscess: Results of an eight-year prospective study of 32,284 operations. Arch Surg 1989; 124:356–361. 52. Muder RR, Brennen C, Wagener MM, Goetz AM. Bacteremia in a long-term-care fa- cility: A five-year prospective study of 163 consecutive episodes. Clin Infect Dis 1992; 14:647–654. 53. Nicolle LE, McIntyre M, Hoban D, Murray D. Bacteremia in a long-term care facil- ity. Can J Infect Dis 1994; 5:130–132. 54. Rudman D, Hontanosas A, Cohen Z, Mattson DE. Clinical correlates of bacteremia in a Veterans Administration Extended Care Facility. J Am Geriatr Soc 1988; 36:726–732. 55. Setia U, Serventi I, Lorenz P. Bacteremia in a long-term care facility. Arch Intern Med 1984; 144:1633–1635. 56. Ackermann RJ, Monroe PW. Bacteremic urinary tract infection in older people. J Am Geriatr Soc 1996; 44:927–933. 57. Bregenzer T, Frei R, Widmer AF, Seiler W, Probst W, Mattarelli G, Zimmerli W. Low risk of bacteremia during catheter replacement in patients with long-term uri- nary catheters. Arch Intern Med 1997; 157:521–525. 58. Steckelberg JM, Melton LJ, Istrup DM, Rouse MS, Wilson WR. Influence on refer- ral bias on the apparent clinical spectrum of infective endocarditis. Am J Med 1990; 88:582–588. 59. Terpenning MS, Buggy BP, Kauffman CA. Infective endocarditis: Clinical features in young and elderly patients. Am J Med 1987; 83:626–634. 60. Selton-Suty C, Hoen B, Grentzinger A, Houplon P, Maignan M, Juilliere Y, Danchin N, Canton P, Cherrier F. Clinical and bacteriological characteristics of infective en- docarditis in the elderly. Heart 1997; 77:260–263. 61. Gagliardi JP, Nettles RE, McCarty DE, Sanders LL, Corey GR, Sexton DJ. Native valve infective endocarditis in elderly and younger adult patients: Comparison of clinical features and outcomes with use of Duke criteria and the Duke Endocarditis Database. Clin Infect Dis 1998; 26:1165–1168. 62. Werner GS, Schulz R, Fuchs JB, Andreas S, Prange H, Ruschewski W, Kreuzer H. Infective endocarditis in the elderly in the era of transesophageal echocardiography: Clinical features and prognosis compared with other patients. Am J Med 1996; 100: 90–97. 63. Wendt C, Krause C, Floss H. Validity of screening procedures for glycopeptide-re- sistant enterococci. Eur J Clin Microbiol Infect Dis 1999; 18:422–427.

428 Mody et al. 64. Cetinkaya Y, Falk P, Mayhall CG. Vancomycin-resistant enterococci. Clin Microbiol Rev 2000; 13:686–707. 65. Leclercq R, Courvalin P. Resistance to glycopeptides in enterococci. Clin Infect Dis 1997; 24:545–556. 66. French GL. Enterococci and vancomycin resistance. Clin Infect Dis 1998; 27(suppl 1):75–83. 67. Levison ME, Mallela S. Increasing antimicrobial resistance: Therapeutic implica- tions for enterococcal infections. Curr Infect Dis Rep 2000; 2:417–423. 68. Morris JG, Roghmann M-C, Schwalbe R. Management of patients with vancomycin resistant enterococci. Infect Dis Clin Pract 2000; 9:10–16. 69. Crossley K. The Long-Term-Care Committee of the Society for Healthcare Epidemi- ology of America. Vancomycin-resistant enterococci in long-term care facilities. In- fect Control Hosp Epidemiol 1998; 19:521–525. 70. Strausbaugh LJ, Crossley KB, Nurse BA, Thrupp LD, SHEA Long-Term Care Com- mittee. Antimicrobial Resistance in long-term-care facilities. Infect Control Hosp Epidemiol 1996; 17:129–140. 71. Garner JS. The Hospital infection Control Practices Advisory Committee, Centers for Disease Control and Prevention. Guidelines for isolation precautions in hospitals. In- fect Control Hosp Epidemiol 1996; 17:53–80. 72. Pugliese G, Weinstein RA, eds. Long-term care, home care, and hemodialysis setting. In: Issues & Controversies in Prevention and Control of VRE: Vancomycin Resistant Enterococci. Chicago, ETNA Communications, 1998:49–70. 73. Armstrong-Evans M, Litt M, McArthur MA, Willey B, Cann D, Liska S, Nusinowitz S, Gould R, Blacklock A, Low DE, McGeer A. Control of transmission of van- comycin-resistant Enterococcus faecium in a long-term-care facility. Infect Control Hosp Epidemiol 1999; 20:312–317. 74. Trick WE, DeMarais PL, Jarvis WR, Tomaska W, Ohlrich S, Hageman J, Rice T, Nathan C, McAllister S, Weinstein RA. Comparison of universal gloving to contact isolation precautions to prevent transmission of multidrug-resistant bacteria in a long- term care facility. Infect Control Hosp Epidemiol 2000; 21:117–121. 75. Hospital Infection Control Practices Advisory Committee (HICPAC). Recommenda- tions for preventing spread of vancomycin resistance. Infect Control Hosp Epidemiol 1995; 16:105–113. 76. Nicolle LE, Bentley DW, Garibaldi RA, Neuhaus EG, Smith PG. SHEA Long-Term- Care Committee. Antimicrobial use in long term care facilities. Infect Control Hosp Epidemiol 1996; 17:119–128. 77. Smith PW, Rusnak PG. Infection prevention and control in the long-term-care facil- ity. Infect Control Hosp Epidemiol 1997; 18:831–849. 78. Jones RN, Marshall SA, Pfaller MA, Wilke WW, Hollis RJ, Erwin ME, Edmond MB, Wenzel RP, and the Scope Hospital Study Group. Nosocomial enterococcal blood stream infections in the SCOPE Program: Antimicrobial resistance, species occur- rence, molecular testing results, and laboratory testing accuracy. Diagn Microbiol In- fect Dis 1997; 29:95–102. 79. Shouten MA, Voss A, Hoogkamp-Korstanje JAA, and the Europeon VRE Study Group. Antimicrobial susceptibility patterns of enterococci causing infections in Eu- rope. Antimicrob Agents Chemother 1999; 43:2542–2546.

24 Gram-Negative Bacteria Vinod K. Dhawan Charles R. Drew University of Medicine and Science, and UCLA School of Medicine, Los Angeles, California I. EPIDEMIOLOGY AND CLINICAL RELEVANCE Gram-negative bacteria are common causes of infection among residents of long- term care facilities (LTCFs). Emergence of antimicrobial resistance in the gram- negative bacteria has been a growing problem in nursing homes, hospitals, and even the community. Antibiotic-resistant organisms may be introduced into nurs- ing homes with the admission of new residents who are already colonized or in- fected. Alternatively, bacterial resistance may emerge in the endogenous flora of residents upon exposure to antimicrobial agents, through either selection of resis- tant strains or spontaneous mutation or gene transfer. There is ample evidence that bacterial resistance negatively impacts the outcome of infections. Data from the Centers for Disease Control and Prevention (CDC) has linked bacterial resistance with higher rates of mortality and morbidity (1). A. Mechanisms of Resistance The development of antimicrobial resistance in microorganisms is a perfect ex- ample of contemporary biological evolution. Over the years, the introduction of new antibiotics has been matched by the development of new mechanisms of re- sistance by the bacteria. Gram-negative bacteria use a variety of strategies to avoid the inhibitory effect of antibiotics and have evolved highly efficient means for dis- semination of resistance traits (Table 1) (see Chapter 21). Among the mechanisms that create resistance to antibiotics in gram-negative bacilli, the production of ␤- lactamase is the single most important factor. ␤-Lactamases are enzymes that hy- drolyze the amide bond in the ␤-lactam ring of the antibiotic, leading to its inac- 429

430 Dhawan Table 1 Mechanisms of Antibiotic Resistance Among Gram-Negative Bacteria Mechanism Antibiotics affected Enzymatic inhibition ␤-lactams, aminoglycosides, chloramphenicol ␤-lactams, aminoglycosides, chloramphenicol, Decreased membrane permeability trimethoprim, sulfonamides Quinolones, tetracyclines Active efflux of antibiotic Aminoglycosides, chloramphenicol, tetracyclines ␤-lactams, quinolones, trimethoprim, sulfonamides Altered ribosomal target Trimethoprim, sulfonamides Altered target enzymes Overproduction of Trimethoprim, sulfonamides target enzymes Bypass of inhibited steps by organisms tivation. The ability of a ␤-lactamase to cause resistance varies with its activity, quantity, and its cellular location within the gram-negative bacteria. A variety of ␤-lactamases, encoded chromosomally or by plasmids (TEM, SHV, or Oxa ␤-lac- tamases) have been described in gram-negative bacteria. The ␤-lactamases have been classified based on their sequences into evolutionary distinct classes A, B, C, Table 2 The Bush-Jacoby-Medeiros Classification of ␤-Lactamases Group Preferred Inhibition by Molecular substrate clavulanate class 1 2a Cephalosporins No C 2b Penicillins Yes A 2be Penicillins, cephalosporins Yes A Penicillins, narrow and extended-spectrum Yes A 2br A 2c cephalosporins, monobactams Diminished A 2d Penicillins Yes D 2e Penicillins, carbenicillin Yes A 2f Penicillins, cloxacillin Yes A 3 Cephalosporins Yes A 4 Penicillins, cephalosporins, carbapenems No B Most ␤-lactams, including carbapenems No Not Penicillins determined Source: Ref. 2.

Gram-Negative Bacteria 431 and D. In addition, a functional classification of ␤-lactamases (group 1, 2, 3, and 4) based on their substrate and inhibitor profiles has been proposed (Table 2). 1. Extended-Spectrum ␤-Lactamases The originally discovered ␤-lactamases (TEM-1, TEM-2, SHV-1) had a rather re- stricted spectrum of activity against antibiotics. More recently, resistance of some gram-negative bacilli to broad-spectrum cephalosporins has been noted to be me- diated by the extended-spectrum ␤-lactamases (ESBLs) designated as Group 2be. Extended-spectrum ␤-lactamases are a group of enzymes that confer resistance to oxyimino cephalosporins (e.g., cefotaxime, ceftazidime, and ceftriaxone) and monobactams. The ESBLs are not capable of hydrolyzing cephamycins and car- bapenems (3). Most ESBLs found in gram-negative bacilli are plasmid-borne variants of the original TEM-1 and SHV-1 enzymes in which one or more amino acid substitutions have expanded the substrate specificity. The ESBLs were first discovered in Europe in 1983 and their prevalence has since increased throughout the world. The ESBLs are most commonly expressed in Klebsiella pneumoniae, K. oxytoca, and Escherichia coli, although they have been detected in other or- ganisms including Salmonella spp, Pseudomonas aeruginosa, Proteus mirabilis, and other Enterobacteriaceae (4–8). Currently, more than 100 of these variants have been described. An updated list of ESBLs is maintained at the website http://www.lahey.org/studies/webt.htm. Different substitutions in ESBLs produce variable effects on the susceptibil- ity of cefotaxime, ceftazidime, and aztreonam to the ␤-lactamases. An emerging mechanism of resistance to ␤-lactamase inhibitors, mediated by the derivatives of TEM and SHV enzymes with a limited number of nucleotide substitutions, has oc- curred in Europe (9). These types of ␤-lactamases have been designated Bush-Ja- coby-Medeiros Group 2br or inhibitor-resistant TEM (IRT) (2). Plasmid-encoded ␤-lactamases can be transmitted among different gram-negative bacteria, resulting in the horizontal spread of antimicrobial resistance. Such plasmids often carry re- sistance to other antibiotics, including tetracyclines, aminoglycosides, chloram- phenicol, trimethoprim, and sulfonamides (8). 2. Inducible Chromosomal ␤-Lactamase AmpC Another important mechanism of resistance in gram-negative organisms is the pro- duction of inducible chromosomal ␤-lactamases, most notably AmpC (Bush-Ja- coby-Medeiros group 1). The presence of a ␤-lactam can cause depression of reg- ulatory genes in these organisms, resulting in ␤-lactamase hyperproduction and inducible resistance to third-generation cephalosporins. Such enzymes are present in 24% to 48% of Enterobacteriaceae strains (10). They have also been noted in some strains of Serratia marcescens, Pseudomonas, Citrobacter, and indole-pos- itive Proteus (11,12). This results in cross-resistance to other ␤-lactams, except

432 Dhawan carbapenems such as imipenem (10) or the fourth-generation cephalosporin, ce- fepime (13). Concomitant aminoglycoside therapy does not prevent the emergence of this resistance (14). The appearance of plasmid-mediated ␤-lactamases similar to AmpC in some strains of K. pneumoniae has resulted in their resistance to cephamycins, oxyimino-␤ lactams, and ␤-lactam inhibitors. The potential plas- mid-mediated transfer has raised concerns about horizontal spread of this resis- tance trait (15). 3. Metallo-Beta-Lactamases Inducible chromosomal enzymes, called metallo-␤-lactamases, confer resistance to ␤-lactam antibiotics in organisms such as Stenotrophomonas maltophilia. Two major functional groups of metallo-␤-lactamases have been identified (16). One group is a set of enzymes with broad substrate specificities capable of hydrolyzing most ␤-lactams, except monobactams. A second group is composed of the “true” carbapenemases—enzymes that exhibit poor hydrolysis of penicillins and cephalosporins. This latter group has been found primarily in Aeromonas spp. Met- allo-lactamases have been recovered from Bacteroides fragilis, S. marcescens, Aeromonas spp, and P. aeruginosa in Japan, but resistance to imipenem has not be- come widespread (16). Carbapenem resistance in K. pneumoniae also can be mod- ulated by plasmid-mediated metallo-␤ lactamase production, raising concerns about widespread dissemination of this resistance mechanism (17). Plasmid-medi- ated resistance to carbapenems is likely to increase and limit the use of these agents as a therapeutic option. 4. Porin Channels Resistance of some gram-negative bacilli to the ␤-lactam antibiotics may also oc- cur through the loss of porin channels in the outer cellular membrane, which de- creases antibiotic entry into periplasmic space. This often leads to increased re- sistance to cephalosporins, cephamycins, and ␤-lactam inhibitors (10,18,19). In P. aeruginosa, carbapenem resistance can occur by mutational loss of a porin chan- nel. Similarly, decreased membrane permeability secondary to porin mutations of- ten leads to quinolone resistance in gram-negative bacilli (20). 5. Efflux Pump Mechanisms A set of multidrug efflux systems in some gram-negative bacteria enables them to survive in a hostile environment (21). The efflux mechanisms pump the antimi- crobial agent out of the cell, preventing its access to the target site. Each efflux pump of gram-negative bacteria consists of three components: the inner mem- brane transporter, the outer membrane channel, and the periplasmic lipoprotein. The molecular mechanism of the drug extrusion across a two-membrane envelope of gram-negative bacteria may involve the formation of the membrane adhesion

Gram-Negative Bacteria 433 sites between the inner and the outer membranes. Quinolone resistance is often modulated by antibiotic efflux systems in addition to alteration in the DNA gy- rase, topoisomerase II and, to a lesser extent, topoisomerase IV (22). 6. Aminoglycoside Resistance Aminoglycoside resistance is modulated by bacterial enzymes, which inactivate the aminoglycoside by variously acetylating, adenylating, or phosphorylating the antibiotic molecule (23). B. Prevalence of Antibiotic-Resistant Gram-Negative Bacteria Most of the information regarding the prevalence of antibiotic-resistant organisms in nursing homes and other LTCFs is derived from surveillance studies of infec- tions or outbreak investigations. No studies have defined the overall magnitude of this problem in a systematic manner. The available data suggest that antibiotic-re- sistant organisms, including gram-negative bacilli, are frequent in the nursing home population (24). Antibiotic resistance among gram-negative bacteria has steadily increased over the years. The antibiotic susceptibility profile of common gram-negative bacilli, as published by the American Society of Clinical Patholo- gists, is presented in Table 3 (25). Current rates of prevalence of ampicillin or amoxicillin resistance in strains of E. coli are about 40% in the United States, 40% to 50% in the United Kingdom and France, 58% in Spain, and 63% in Israel (26). Ampicillin-resistant isolates of E. coli and cephalothin-resistant isolates of Kleb- siella spp are common in nursing homes (24). Aminoglycoside resistance has been noted in a significant proportion of uropathogens isolated from nursing home residents (27). In one study 33% of such organisms were noted to be resistant to gentamicin (28). Another study detected colonization of urine or perineum with trimethoprim-resistant gram-negative bacilli in 52% of residents in a Department of Veterans Affairs (VA) nursing home (29). Resistance of gram-negative bacilli to fluoroquinolones has also been described. Prospective surveillance in seven skilled nursing facilities in southern California found about a third of the urinary Pseudomonas isolates and 12% of isolates of the family Enterobacteriaceae were resistant to norfloxacin (30). Extended-spectrum ␤-lactamase-producing organisms, which are being identified worldwide (Table 4), are probably more prevalent than currently rec- ognized because they are often undetected by routine susceptibility testing meth- ods (31). These organisms are commonly encountered in nosocomial infections and have been implicated in several nursing home outbreaks (32,33). The preva- lence of ESBL-producing K. pneumoniae is on the incline with rates approaching as high as 40% in some hospitals in the northeast United States (34). As reported by the CDC, in 1998, the rate of ceftazidime-resistant K. pneumoniae reached 10.7% and ceftazidime-resistant E. coli reached 3.2% in intensive care units