| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Pharmacy, Deaconess Medical Center, Washington State University, Spokane, Washington
b Department of Pharmacy, Palomar Medical Center, Escondido, California
c Department of Surgery, Palomar Medical Center, Escondido, California
* Address correspondence to Dr Kruse, Palomar Medical Center, Department of Pharmacy, 555 E Valley Parkway, Escondido, CA 92025 (Email: michael.kruse{at}pph.org).
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsCommentAcknowledgmentsReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsCommentAcknowledgmentsReferences |
|---|
After cardiothoracic surgery, the incidence of surgical-site infections (SSIs) ranges from 0.5% to 14.3% [2]. Postsurgical mediastinitis occurs in 1% to 4% of patients after coronary artery bypass graft surgery [3, 4], of which up to 65% are caused by MRSA [5]. This is a likely reflection of the documented 0.4% to 20.6% of nasal MRSA colonization [6–8]. Compared with methicillin-sensitive S. aureus (MSSA) mediastinitis, MRSA mediastinitis has up to an 11-fold increased mortality rate, and is associated with approximately $35,000 to $40,000 in excess charges per infection [4, 9].
Methicillin-resistant S. aureus infection is at the center of a paradigm shift in medicine that includes legislative, reimbursement, public reporting, and regulatory changes. A number of states (Illinois, New Jersey, Pennsylvania, and California) and the Veterans Health Administration mandate MRSA surveillance in variously defined patient populations. In December 2007, a federal bill, S. 2525 MRSA Infection Prevention and Patient Protection Act, was introduced for consideration, which would mandate MRSA screening in all US hospitals by no later than January 1, 2012 [10]. As of October 1, 2008, the Centers for Medicare and Medicaid Services (CMS) no longer reimburses for post–coronary artery bypass graft mediastinitis during the same hospitalization as the surgery [11]. This will not affect reimbursement for discharged patients who subsequently are admitted with infection. However, pay-for-performance for other nosocomial infections will force hospitals to consider previously cost-prohibitive screening efforts. In October 2008, the Joint Commission joined the American Hospital Association, Society of Healthcare Epidemiology in America, the Infectious Diseases Society of America, and the Association for Professionals in Infection Control and Epidemiology in collating infection prevention guidelines: the Compendium of Strategies to Prevent Healthcare-Associated Infections in Acute Care Hospitals [12]. Although numerous infection control organizations have published previous guidelines, this marks the Joint Commission's first involvement. It is not yet clear whether this may predict additional accreditation oversight of nosocomial infection rates.
The Compendium guidelines replace or supplement previous guidelines and help hospitals determine the role of each intervention in their infection control practices. Overall, the aforementioned paradigm shift will likely cause a more rapid expansion of active MRSA surveillance in hospitals. Active MRSA surveillance consists of MRSA screening, isolation, contact precautions, and decolonization. Existing Society of Thoracic Surgery (STS) guidelines presuppose complete active surveillance may not be possible in most hospitals. The implementation of active surveillance, whether targeted in a certain patient population or universal in all patients, is controversial. Whether or not active surveillance decreases the incidence of MRSA nosocomial infections or MRSA mediastinitis remains unclear. This article will review the Compendium and STS guidelines and highlight the benefits and concerns of implementing active MRSA surveillance in cardiothoracic surgery patients.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsCommentAcknowledgmentsReferences |
|---|
Laboratory Methods for Methicillin-Resistant Staphylococcus aureus Screening
Existing guidelines do not recommend any specific laboratory method to perform MRSA screening. Institutions performing active surveillance primarily use either selective differential media, of which there are three—CHROMagar MRSA [13], Spectra MRSA, and MRSA Select—or real-time polymerase chain reaction (PCR) analysis, of which two are US Food and Drug Administration–approved: BD GeneOhm [14] and Cepheid GeneXpert. According to the Compendium guidelines, the selection of a particular MRSA screening method should be based on numerous elements including sensitivity and specificity, turnaround time, laboratory capabilities, number of specimens processed, and cost-benefit calculations [12]. Table 1
compares the various MRSA laboratory-screening methods an institution may consider.
|
Methicillin-Resistant Staphylococcus aureus Screening Culture Sites
Various body sites may be cultured to determine MRSA colonization status. The Compendium guidelines do not specify particular screening sites but note that the anterior nares appear to be positive most frequently [12]. The most common sites for MRSA colonization are the anterior nares (68% to 88%), throat (53%), perianal area (53%), groin (49% to 50%), umbilicus (56%), and axillae (31%) [7, 20]. However, it must be noted that selective differential media and PCR are only US Food and Drug Administration–approved for nasal swabs, with PCR GeneXpert MRSA just recently US Food and Drug Administration–approved for skin and soft tissue samples in September 2008.
Adding throat swabs to nasal swabs has demonstrated an increased MRSA yield of 12.5% to 22% [21, 22]. Although screening additional sites increases the yield, the clinical advantage and cost-benefit of swabbing added sites is unknown. The anterior nares have the highest yield, and may be colonized for a mean of 42 months [23]. Although pooling samples could minimize the cost and labor associated with including additional swab sites, performing only nasal swabs is the most convenient for health-care professionals and the least invasive for patients.
Methicillin-Resistant Staphylococcus aureus Colonization and Inpatient Methicillin-Resistant Staphylococcus aureus Prevalence
The prevalence of MRSA nasal colonization in the United States significantly increased from 0.8% to 1.5% in 2004 [8], and varies per institution. One would expect MRSA colonization to be higher in facilities with increased inpatient MRSA prevalence. However, the incidence of invasive MRSA infections is not always associated with the incidence of community-associated MRSA [24].
Because the prevalence of inpatient MRSA is not predictive of the prevalence of MRSA colonization, it is invaluable to determine the baseline incidence of MRSA colonization. Once colonized or infected, MRSA colonization may persist more than 3 years after hospital discharge [25]. Patients colonized with MRSA within 1 year have demonstrated an increased risk for MRSA infection and mortality, with the greatest risk of infection during the first quarter (p < 0.001). Greater than 10% of newly colonized inpatients exhibit MRSA infection during the same hospital stay, and newly colonized intensive care unit (ICU) patients establish up to a 40% risk of short-term MRSA bacteremia. Compared to those with prior MRSA infection who received anti-MRSA therapy, patients with prior MRSA colonization obtained significantly more subsequent MRSA infections [23].
Methicillin-Resistant Staphylococcus aureus Colonization and Surgical Outcomes
Methicillin-resistant S. aureus colonization has been associated with increased rates of infection and worsened outcomes. Various studies have demonstrated that active surveillance identifies up to 91% of MRSA colonization undetected by clinical culture alone. Only 33% of these patients are identified with MRSA during their stay through clinical cultures [26, 27]. Limited research has evaluated the correlation between presurgical MRSA colonization and postoperative infection.
In a prospective, interventional cohort study by Robicsek and associates [28], the introduction of PCR-based universal surveillance using nasal screening was associated with a large reduction in MRSA infection during admission and 30 days after discharge. Through isolation and optional decolonization, ICU surveillance demonstrated a positive trend in decreasing MRSA infections by 36.2%, whereas universal surveillance showed a significant decline in MRSA infections by 69.2% from baseline.
In contrast, a prospective cohort study among 21,754 surgical patients, by Harbarth and associates [29] detected no significant reduction in nosocomial MRSA infections after implementing PCR-based universal surveillance in surgical wards. Cultures of the anterior nares, perineal region, and other sites (catheter insertion sites, skin lesions, urine) were obtained when clinically indicated. Patients colonized with MRSA (5.1%) were isolated and decolonized with nasal mupirocin and chlorhexidine body wash for 5 days. Potential explanations for the unimproved incidence of nosocomial MRSA infections are the low prevalence of MRSA bacteremia or the delay in the positive MRSA results. Thirty-four percent of patients did not receive appropriate vancomycin prophylaxis and acquired SSIs [29]. If all MRSA-colonized patients had received appropriate surgical prophylaxis, active surveillance in surgical patients may have demonstrated a significant benefit.
Whether or not active surveillance is beneficial in cardiac surgery patients is not clearly defined. A study by San Juan and associates [30] suggests little benefit of active surveillance in preventing MRSA mediastinitis. Preoperative nasal cultures indicated 15.5% MSSA colonization and 0.4% MRSA colonization. Poststernotomy mediastinitis was diagnosed in 1.2% of patients (17 of 1,432), of which 47% were MRSA and 53% were MSSA. Although MSSA mediastinitis was correlated with preoperative MSSA colonization (p < 0.00001), MRSA mediastinitis was not associated with preoperative MRSA colonization. Therefore, although preoperative decolonization may be an adequate means of prevention for MSSA mediastinitis, postoperative infection control measures seem essential in preventing MRSA mediastinitis. These results may have been related to the low rate of preoperative MRSA colonization or the possibility that MRSA was obtained postoperatively. In cardiac surgery patients with a higher prevalence of 2.5% MRSA colonization, active surveillance was associated with decreased SSIs [31]. After a 12-month implementation of PCR active surveillance, decolonization, and a single prophylactic dose of teicoplanin, a decrease in total SSIs of 3.3% to 2.22% and sternal SSIs of 2.87% to 1.83% in cardiac surgery patients was observed. Although this did not reach statistical significance, the results are promising in terms of reducing SSIs after cardiac surgery.
Various active surveillance studies have indicated that MRSA colonization poses an inherent risk for increased infection and mortality. Although limited, the existing data suggest that active surveillance may prevent MRSA SSIs when MRSA colonization is at least 2.5%. In the absence of active surveillance, only 17.8% to 33.3% of isolation days are identified [28]. Further studies are needed to delineate whether or not MRSA colonization is directly correlated with MRSA SSIs, and determine the additional benefit of active surveillance in cardiothoracic surgery patients.
Antibiotic Prophylaxis
Regardless of colonization status, antibiotic surgical prophylaxis is a mainstay of infection prevention. However, not all facilities have revised their antibiotic practices despite evidence showing potential to decrease SSIs [32]. A surgeon should review guidelines when evaluating infection control measures. The most recent consensus guidelines for antibiotic prophylaxis come from the CMS and Centers for Disease Control and Prevention as part of the National Surgical Infection Prevention Project [33]. These have been folded into the Surgical Care Improvement Project and CMS pay-for-performance. In fiscal years 2007 and 2008, CMS required reporting of antibiotic timing and selection, respectively, for hospitals to receive full payment [34]. The Compendium guidelines discuss antibiotic selection and timing but do not provide a detailed discussion [12]. The 2007 STS guidelines for antibiotic choice in cardiac surgery fulfill both CMS requirements and Compendium guidelines [35].
Antibiotic selection
Cefazolin is the standard of care for prophylaxis in cardiovascular surgery [33, 35]. The routine use of vancomycin is not recommended, but guidelines recognize the potential need for vancomycin in cardiovascular surgery when there is a high endemic incidence of MRSA or patients are high-risk [12]. A single dose of vancomycin 15 mg/kg must be infused for 1 hour and be completed within 1 hour of skin incision. It is unclear whether dual therapy should be recommended, but two problems may exist if cefazolin is excluded—loss of cefazolin's gram-negative coverage and poor activity against MSSA. Therefore, guidelines mention combining vancomycin with cefazolin.
An ideal regimen has not been studied for penicillin-allergic patients in the setting of high MRSA incidence. Patients with a severe penicillin allergy (hives, anaphylaxis, or angioedema) may be given vancomycin without cefazolin [35]. Clindamycin should not be used with vancomycin because vancomycin's activity may be compromised [36]. Cefazolin's gram-negative and MSSA coverage can be replaced with gentamicin 4 mg/kg or levofloxacin if the patient has renal compromise.
Dose adjustment for weight and readministration
Two issues that may be overlooked include readministration during the case and weight-adjusted dosing of cefazolin. Guidelines recommend a second dose of 1 g every 3 to 4 hours and recommend cefazolin 2 g rather than 1 g in obese patients [33, 35]. The National Surgical Infection Prevention guidelines use a cutoff of 80 kg, and the STS guidelines use 60 kg. Studies of morbidly obese patients are used to support this recommendation. A pharmacokinetic study has not been completed to determine whether to use 60 kg or 80 kg for dose selection.
Infusion time is a practical issue with antibiotic selection. Cefazolin can be given during a few minutes in the operating room. Vancomycin and levofloxacin require initiation in the preoperative area as rapid infusion can lead to hypotension or cardiac arrest.
Antibiotic stewardship program
In addition to traditional infection control measures, a robust antibiotic stewardship team is recommended by the Society of Healthcare Epidemiology in America and the Infectious Diseases Society of America [37]. At Palomar Medical Center, an infectious diseases pharmacist was hired to work with the infectious diseases physician. In the first 2 years, the MRSA prevalence decreased from 60% to 42%, and antibiotic acquisition costs decreased by more than $200,000 per year. Surgeons in any specialty should support or encourage the hospital administration in these efforts.
Decolonization
Practical considerations related to MRSA decolonization include patient selection, type of surgery, selection of agent, and local resistance patterns.
Patient selection: Targeted versus population
Guidelines recommend against decolonization except when using an evidence-based approach for infection prevention, psychological benefit of the patient, or cost benefit [35].
The Society of Thoracic Surgeons guidelines recommend routine mupirocin administration for all patients undergoing cardiac surgical procedures without confirmation of negative Staphylococcal colonization [35]. The guidelines acknowledge that limiting mupirocin to colonized patients "would appear to be a sensible approach." However, with limited access to PCR testing, a broad population-based approach is recommended. Limiting the population-based approach to high-risk populations is consistent with Compendium guidelines that recommend against widespread or prolonged use. If surgeons influence the selection of the PCR device in an institution, rapid identification of patients may limit mupirocin decolonization to MRSA-colonized patients.
Mupirocin
Numerous studies support the use of intranasal mupirocin for decolonization. However, few studies used controls or randomization. Mupirocin regimens varied in dosing frequency, number of days, and day of initiation. Since the 2007 STS guidelines, a meta-analysis and a review have examined the role of mupirocin in surgical patients.
A meta-analysis by van Rijen and colleagues [38] pooled four studies to assess mupirocin treatment in S. aureus nasal carriers. Three of the four studies included cardiovascular patients—two studies were of solely cardiovascular patients. The meta-analysis excluded trials that used a historical control or did not have control groups and randomization. Overall, there was a significant reduction in postoperative S. aureus infection (relative risk, 0.55; 95% confidence interval, 0.34 to 0.89) and a trend toward a decreased incidence of SSIs, but this was not statistically significant (relative risk, 0.64; 95% confidence interval, 0.38 to 1.06).
The review by Trautmann and associates [39] had similar exclusion criteria to van Rijen and colleagues, but historical controls were allowed. In contrast to the meta-analysis [38], the review allowed patients wwho were not colonized with MRSA. Eleven trials were identified. Four used randomization, and seven used historical controls. Only four studies reduced overall SSI, and two reduced SSI by S. aureus. Trautmann and coworkers [39] recommend against routine decolonization in patients with unknown colonization status. This recommendation conflicts with STS guidelines. However, on further review, three of the four studies with decreased SSIs studied a cardiovascular population. Therefore, these three studies would support STS guidelines in recommending mupirocin decolonization when status is unknown.
Chlorhexidine
Chlorhexidine utilization may confound the study of mupirocin or other measures [40]. It is widely used for skin cleansing, decolonization, and impregnated dressing and catheters. Six of the studies in the review by Trautmann and colleagues [39] described concomitant use of chlorhexidine showers. Despite its success in reducing microbial burden, a cause-and-effect relationship in reducing SSIs has not been found [40]. In cardiovascular surgery, chlorhexidine has been investigated for both oral decontamination and nasal decolonization. DeRiso and associates [41] randomized 353 consecutive open heart patients to an oral 0.12% rinse or placebo. Patients received the oral rinse preoperatively and twice daily postoperatively until discharge from the ICU. Nosocomial and respiratory tract infections decreased by 65% and 69% respectively. Nonprophylactic intravenous antibiotics were reduced by 43%, and mortality was reduced (1.16% versus 5.56%). Segers and colleagues [42] reported similar results using an oral rinse applied four times daily plus nasal chlorhexidine ointment four times daily. Lower respiratory tract infections and deep SSIs were reduced. The Compendium guidelines now recommend routine chlorhexidine bathing in ICU patients as a replacement for soap and water [12]. This may further complicate determining the treatment effect of other agents, but future studies versus mupirocin may be of interest.
Systemic treatments
Various agents and regimens have been used for oral systemic therapy. These include systemic rifampin, doxycycline, sulfamethoxazole-trimethoprim, ciprofloxacin, clindamycin, and minocycline [43]. Although these regimens may be effective, especially in decolonization of known carriers, a much higher incidence of resistance may be reported after therapy. A review of seven regimens including rifampin resulted in up to 40% resistance after therapy.
Mupirocin and chlorhexidine resistance
Compendium guidelines recommend against widespread mupirocin use but recognize a role in cardiothoracic surgery. Miller and colleagues [44] reported the danger of widespread use beyond high-risk populations. In a public teaching hospital, mupirocin was limited for 2 years to MRSA-infected and colonized patients. In the third year, all hospitalized patients received mupirocin. Mupirocin resistance increased from 2.7% to 65%. A limitation of this observation was the use of mupirocin for the full hospital stay. Most decolonization regimens in other studies have used no more than 5 days of therapy. Guidelines still recommend occasional susceptibility testing when mupirocin is used in any population of patients.
Limited data exist regarding resistance to chlorhexidine [40]. Although the solutions used for topical application may be above any minimum inhibitory concentration for resistant organisms, theoretical cross-resistance to other antibiotics may be induced [45]. Vali and colleagues [45] isolated 120 MRSA strains for surface disinfection with chlorhexidine. In some strains, postexposure minimal inhibitory concentrations were higher for chlorhexidine, cefotaxime, vancomycin, gentamicin, cefuroxime, and oxacillin.
Cost-Effectiveness of Active Surveillance
The cost of active surveillance includes screening materials, nursing staff to perform swabs, laboratory testing, contact isolation materials, private rooms, and potentially, decolonization antimicrobials. Various studies have investigated the cost-savings of active surveillance in minimizing MRSA infections. In analyzing cost, the currency used and inflation must be considered.
Cost of an active surveillance program
The method of screening, population screened, and body sites swabbed determine the cost of an active surveillance program. Cost reporting methods differ among studies and should be taken into account, some reporting costs in Canadian or European money and others reporting per patient, day, month, or year.
According to a Canadian simulation model, approximately 1,095 screening tests per year would be conducted in a 34-bed ward assuming three admissions per day. The estimated annual total cost for culture-based screening was C$5,517, which was cost-effective if one MRSA infection was prevented every 2 to 3 years (1 infection per 24,000 to 36,000 patient-days) [46]. When screening only high-risk patients (anterior nares, perineum, skin lesions, and catheter exit sites), the cost of the culture-based surveillance program was substantially higher at C$54,901 per year [47].
Costs of implementing active surveillance in ICUs were significantly higher. In studies, ICU patients were screened on admission or transfer, and weekly thereafter. The estimated total cost of culture-based ICU active surveillance, including costs of laboratory tests, nursing time to perform cultures, and the number of screening cultures performed, was $113,955 ($54,381 for cultures, $58,573 for contact isolation) for 16 months [48], approximately $7,000 per month, versus $3,475 per month for only screening and isolation [26]. When the incidence of MRSA colonization on admission to the ICU was 1% to 7%, active surveillance was cost-effective in high-risk settings (greater than 25% risk for infection). Methicillin-resistant S. aureus screening, along with contact precautions (ie, single room isolation, wearing gown and gloves, handwashing) for MRSA colonized and infected patients cost $340 to $1,480 per patient [49].
The cost of PCR-based active surveillance can be substantially higher when the cost of the equipment and reagents are included. Compared with the C$5,517 cost for culture-based screening, the total cost for PCR screening was C$38,325. Despite its excess cost, PCR screening was cost-effective if three to four MRSA infections were prevented each year (1 infection per 3,000 to 4,000 patient-days) [46], an equivalent savings of approximately C$32,000 per year.
Cost of preventing methicillin-resistant staphylococcus aureus surgical-site infections
The average attributable cost of an MRSA infection in the United States has been estimated to be between $7,781 to $34,000 per infection [47]. Compared with patients with MSSA SSIs, patients with MRSA SSIs had a significant 1.19-fold increase in hospital charges and mean attributable excess charges of $13,901 per SSI. Each MRSA SSI was estimated to cost $118,415. Preventing MRSA SSIs has saved institutions from $240,289 to $447,720 per year [48, 50] and is considered cost-effective if at least four infections (11% of MRSA infections) or two SSIs were prevented per year [51].
Although active surveillance may be costly to implement, studies support that it is a cost-effective investment to minimize SSIs such as mediastinitis after coronary artery bypass graft surgery, which is no longer reimbursable by CMS. For the 2006 fiscal year, there were 108 reported cases of mediastinitis after coronary artery bypass graft surgery in Medicare patients, with average charges for hospital stays equal to $304,747 [11]. Even if the most expensive active surveillance program cited in studies were implemented at $262,554 per year [47], there would still be a significant annual cost savings of approximately $40,000.
Confounding Factors in Methicillin-Resistant Staphylococcus aureus Infection
Environmental and health-care worker colonization
In a surgical ward, 10.7% of environmental sites may be MRSA contaminated [52]. Furniture (ie, bed, bedside rails, bedside table, toilets), nurse call buttons, patient room computers, plastic patient charts, medical equipment (ie, infusion pumps, ventilators), and the floor rank among the top colonizers of MRSA in a patient's room [10, 52, 53]. A study by Wilson and associates [54] reported that MRSA transmission from the environment to the patient was unlikely because the strains patients obtained differed from those found in bed areas. However, a number of studies have demonstrated environmental transfer of MRSA, and until conclusive results are available, the environment should be considered a potential source of MRSA.
In addition to the environment, health-care workers may also spread MRSA. About 5% of health-care workers become colonized with MRSA and are usually transient carriers [55]. The highest rate of nasal MRSA carriage was reported in surgical units and the operating room, accounting for 68% of MRSA—2.3-fold higher than other units [56]. Based on these data, one might think that MRSA colonization in nurses and surgeons should be eradicated. The Compendium guidelines do not recommend routine screening of health-care workers unless directly associated with an outbreak [26]. In contrast, in some European countries, since 1992, active surveillance includes the screening and decolonization of both patients and health-care workers. Colonized health-care workers are removed from patient care for at least 48 hours after initiation of decolonization therapy or proven eradication [57]. The concern is not only the additional cost but also the potential stigmatization of the colonized staff.
If environmental or health-care worker MRSA colonization is significant, the implementation of active surveillance or any attempt to eradicate MRSA from patients may be compromised. As a result of inadequate contact precautions, it has been noted that patients colonized or infected with MRSA are 16 times more likely to transmit MRSA [25]. Therefore, the Compendium guidelines emphasize the importance of basic infection control measures such as hand hygiene and contact precautions before considering a more robust program that consists of active surveillance [12]. Surveillance is recommended in at least high-risk populations once compliance of basic infection control practices is verified.
|
Comment |
|---|
|
TopAbstractIntroductionMethodsCommentAcknowledgmentsReferences |
|---|
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMethodsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsCommentAcknowledgmentsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
