Despite advances in antimicrobial therapy and improvements in supportive care and prevention, methicillin-resistant Staphylococcus aureus (MRSA) infections remain a significant challenge in the intensive care unit (ICU). Such infections are significant causes of morbidity, mortality, and increased patient cost. Of the 2 million nosocomial infections acquired annually in the United States, approximately 60% involve antibiotic-resistant bacteria (defined as resistant to 1 or more agents).1 Overall, more than 50% of nosocomial S. aureus infections are methicillin-resistant, with associated treatment costs ranging from $100 million to $30 billion, based on various estimates.1,2 Furthermore, such infections are increasing in prevalence.3 For these reasons, the American Thoracic Society (ATS) and the Infectious Diseases Society of America (IDSA) recently developed new guidelines for the management of nosocomial pneumonia caused or exacerbated by MRSA; however, MRSA can also invade other anatomic sites.4
The MRSA infections typically seen in the ICU are pneumonias, bacteremic infections, and infections of traumatic and surgical wounds. Because of their virulence, much of the study of MRSA infections is focused on the nosocomial pneumonias. These are categorized as hospital-acquired pneumonia (HAP), which includes 2 subtypes: ventilator-associated pneumonia (VAP) and healthcare-associated pneumonia (HCAP).4
The most common route for the acquisition of bacteremic infections is the I.V. catheter,5,6 with MRSA one of the most common causes of catheter-related bloodstream infections (CR-BSIs). Although such infections are generally acquired nosocomially, they can also be acquired in the community7; for example, MRSA infections of postoperative wounds are often seen among patients admitted from nursing homes to acute care facilities.8 Common risk factors for MRSA infections are listed in Table 1.
MRSA is often resistant to antibiotics outside the β-lactam class. As a result, some strains may be referred to as multidrug-resistant MRSA, or MDR MRSA. Although other antibiotic-resistant pathogens exist, MDR MRSA is the most common,3 and its prevalence is increasing. HAP caused by MDR MRSA, for example, affects 25% of all ICU patients and accounts for 50% of prescriptions for antibiotics. VAP occurs in 9% to 27% of all intubated patients; HCAP is often secondary to viral influenza.4
Several factors increase a patient’s risk for acquiring a respiratory infection with MDR MRSA, including the following4:
• having received antimicrobial therapy within the preceding 90 days
• being hospitalized for >5 days
• high rates of antibiotic resistance in the community at large or within a specific hospital unit
• having known risk factors for HCAP
• having an immunosuppressive disease or receiving immunosuppressive therapy.
Among patients with trauma, head injuries, abdominal injuries, ICU stays of more than 7 days, and treatment with glycopeptide antimicrobials are associated with MDR MRSA infection.9 In addition to MDR MRSA, other strains of drug-resistant S. aureus are emerging as clinical concerns. These include glycopeptide–intermediately resistant S. aureus and vancomycin-resistant S. aureus (VRSA). Although rare, infections caused by these organisms are associated with high rates of morbidity and mortality.10,11
Various investigators have reported similar findings regarding length of hospital stay and economic costs related to resistant infections. For example, in 1999, Heyland and colleagues demonstrated that patients with VAP are 5.8% more likely to die than are comparable intubated patients without VAP12; infected patients remained in the ICU on average 4.3 days longer than those who did not contract MDR MRSA infection. Cosgrove et al found that patients infected with MRSA had hospital stays 30% longer and hospital charges 35% higher than those of patients with infections that responded to methicillin.13 Others have reported similar findings.14
The ATS–IDSA guidelines published in 2005 are an updated version of those issued by the same organizations in 1996.4 As stated earlier, the updated guidelines recognize new subcategories of HAP (VAP and HCAP) and include information about their epidemiology and pathogenesis, in addition to the modifiable risk factors associated with them. HAP, for example, is defined as a pneumonia occurring 48 hours or more after admission in a patient who did not have pneumonia at the time of admission; VAP refers to pneumonia developing more than 48 to 72 hours after endotracheal intubation.
HCAP is described as pneumonia developing in a patient in any of the following circumstances4:
• hospitalized in an acute care setting for >2 days within 90 days of the current infection
• residing in a nursing home or long-term care facility
• receiving I.V. antibiotic therapy, chemotherapy, or wound care within 30 days of the current infection
• receiving outpatient care in a hospital or hemodialysis clinic.
The subtypes of HAP are also differentiated according to the type of bacterial infection involved. Among gram-negative bacilli, common underlying pathogens include Pseudomonas aeruginosa, Escherichia coli, Klebsiella pneumoniae, and Acinetobacter. Among gram-positive bacteria, S. aureus (eg, MRSA) is common, particularly in patients with diabetes mellitus, with head trauma, or admitted to an ICU.4
The ATS–IDSA guidelines note that a definitive diagnosis of HAP is often challenging. The guidelines identify 10 points that should be addressed before a diagnosis of HAP is made (Table 2).4
These steps address several ambiguities in the diagnosis of HAP. For example, if HAP is suspected, a chest X-ray is typically ordered. If the X-ray shows a new or progressive infiltrate along with clinical findings suggesting infection, HAP may be present but is not confirmed. Other suspect clinical findings include a new onset of fever, purulent sputum, leukocytosis, and a decline in oxygenation. However, a diagnosis of tracheobronchitis may be appropriate if the patient has fever, leukocytosis, purulent sputum, and a positive culture of either sputum or a tracheal aspirate but no new lung infiltrate.
In mechanically ventilated patients, nosocomial tracheobronchitis is linked to a longer ICU stay and longer periods of mechanical ventilation, but not to increased mortality. However, because antibiotic treatment may prevent subsequent pneumonia, which is linked to increased mortality, physicians may want to consider antibiotic treatment in such patients, according to the guidelines.4
The updated guidelines identify several modifiable risk factors for HAP, including the following4:
• intubation, particularly nasotracheal and nasogastric intubation, and mechanical ventilation
• recumbent versus semirecumbent positioning
• enteral nutrition
• oropharyngeal colonization
• prophylactic agents to prevent stress bleeding.
Several actions are recommended to modify these risks. These include vigorous infection control measures and surveillance of ICU infections, avoidance of intubation if possible, and use of noninvasive ventilation when possible. Recommended measures also include using orotracheal and orogastric tubes instead of nasotracheal and nasogastric tubes, placing the patient in a semirecumbent position (30-40 degrees), and in some cases administering oral antiseptics and oral antibiotics to reduce the risk for the migration of colonization. Leukocyte-depleted transfusions of red blood cells and rigorous glucose control in patients with diabetes should also be considered.4
Educating staff about these modifiable risk factors for VAP is effective in reducing its incidence.15 In one study involving 4 hospitals, investigators found that VAP rates dropped an average of 46% after implementation of an education intervention designed to prevent VAP. In the year before the intervention, the rate was 8.75 per 1,000 ventilator days, whereas during the 18-month period following the intervention, the VAP rate dropped to 4.74 per 1,000 ventilator days (P<0.001). In a community hospital where the rate of compliance with the educational protocol was lowest, however, the investigators observed no change in VAP incidence rates.15
The ATS–IDSA guidelines stress the importance of promptly administering empiric antibiotic therapy—because of the risk for excessive mortality associated with delayed antimicrobial treatment—and of screening patients for risks for MRSA and other MDR organisms. Patients with infection beginning within the first 4 days after hospitalization are to be considered at high risk, as are those transferred from nursing homes or cared for at facilities such as dialysis centers. According to an algorithm developed in the guidelines, patients at low risk for HAP, VAP, and HCAP can be treated with limited-spectrum antibiotics, but high-risk patients should be treated with broad-spectrum agents (Figure).4 Limited-spectrum agents include ceftriaxone, levofloxacin, moxifloxacin, ciprofloxacin, ampicillin-sulbactam, and ertapenem. Broad-spectrum agents include cefepime, ceftazidime, imipenem, meropenem, piperacillin-tazobactam, amikacin, gentamicin, tobramycin, linezolid, and vancomycin.
Combination therapy is recommended for patients at high risk for infection with MDR and MRSA pathogens. This approach typically consists of 3 antibiotic agents. The guidelines recommend the following combinations: an antipseudomonal cephalosporin or carbapenem or a ß-lactam with a ß-lactamase inhibitor, plus an antipseudomonal fluoroquinolone or an aminoglycoside, plus linezolid or vancomycin. In cases of MRSA infection, regardless of the infection site, adequate tissue penetration is critical for patients to benefit from treatment. With linezolid, for example, I.V. dosing of 600 mg every 12 hours is considered necessary in adults for adequate antibiotic penetration of lung tissue.16 With vancomycin, the comparable recommended dosage is 15 mg/kg every 12 hours in patients with normal renal function.4
Adequate antibiotic dosing for an appropriate length of time is essential to the successful treatment of patients at risk for MDR infections. Table 3 shows the recommended I.V. doses of antibiotics used to treat adult patients with HAP who are risk for MDR infections.4
The appropriate duration of therapy depends on the nature of the infection and whether the patient has shown an adequate initial response to therapy. In patients with a good initial response to treatment and in whom clinical features of the infection have resolved, the guidelines recommend shortening the duration of treatment from the typical 14 to 21 days to 7 days. This recommendation is based on findings that longer treatment is associated with recurrent colonization by resistant organisms. In patients with P. aeruginosa infection, however, the recommended duration of treatment is still 14 to 21 days.
After empiric treatment has begun, a response is typically seen within 48 to 72 hours. During that time, no changes in therapy are recommended unless the patient’s condition is deteriorating. After day 3, if the patient has not responded, the clinician should reassess the clinical parameters. If the patient is responding, the clinician should consider the culture data obtained before antibiotic therapy was begun and de-escalate the antibiotics by using the narrowest possible treatment regimen based on the culture data. Proponents of de-escalation define it as a method of providing appropriate initial treatment to patients with serious bacterial infections while avoiding the improper use and/or overuse of antibacterial agents. The goal of this approach is to prevent the development of resistance.17
A few studies have linked the prevention of protein synthesis to the effective treatment of infection with MDR organisms, although further study is needed. Blocking protein synthesis in turn inhibits the production of microbial toxin. An important in vitro proof-of-concept study demonstrated that linezolid inhibits the synthesis of proteins that are essential to cell division in several targeted organisms, including E. coli.18 Quinupristin-dalfopristin has demonstrated the ability to impair bacterial protein synthesis at both the early step of peptide chain elongation and the late step of peptide chain extrusion and has bacteriostatic activity against vancomycin-resistant Enterococcus faecium.19 Several classes of antibiotics—including the macrolides, tetracyclines, aminoglycosides, and oxazolidinones—have demonstrated the ability to inhibit protein synthesis.
The ATS–IDSA guidelines call for treating all patients who have nosocomial pneumonia with I.V. formulations of antibiotics initially. Depending on the treatment regimen, physicians can then change to oral or enteral formulations if the patients are responding and the functioning of their gastrointestinal tract is intact.4 As examples, quinolones—such as levofloxacin—and linezolid are produced in I.V. formulations with bioavailability equivalent to that of their oral counterparts. Switching to oral therapy can reduce the costs associated with the preparation and administration of I.V. drugs and a prolonged hospital stay. Oral therapy also reduces the inconvenience and risk for new infection associated with an indwelling catheter and extended hospital stay.
Because of the serious nature of infection with MDR pathogens, appropriate management is critical, both for the sake of affected patients and to prevent the spread of infection. Therefore, the new ATS–IDSA guidelines for managing all subtypes of HAP will have a far-reaching effect and require the attention of those who treat patients critically ill with pulmonary and/or infectious disease. Furthermore, because several of the MDR pathogens affect organ systems such as the skin and surrounding structures, as well as the deep organs and circulatory system, the management approaches listed in the guidelines may influence the practice of clinicians beyond those involved in respiratory care. They may be useful to diabetes experts and those who treat patients susceptible to cutaneous infections, surgeons who treat deep wound infections, and providers of care to patients affected with immunosuppression.