Ventilator-associated pneumonia (VAP) is the most frequent intensive care unit (ICU)–acquired infection among patients who are treated with mechanical ventilation. In contrast to infections of other organs (e.g., urinary tract and skin), for which mortality ranges from 1% to 4%, the mortality rate for VAP, defined as pneumonia occurring more than 48 hours after the onset of mechanical ventilation, ranges from 20% to 50%, and can even be higher when lung infection is caused by high-risk pathogens. Although the attributable mortality rate for VAP is still debated, good evidence indicates that VAP prolongs the duration of mechanical ventilation and of the ICU stay. Approximately 50% of all antibiotics prescribed in an ICU are administered for respiratory tract infections. Because several studies have shown that appropriate antimicrobial treatment of patients with VAP significantly improves outcome, rapid identification of infected patients and accurate selection of antimicrobial agents are important clinical goals. Agreement is lacking, however, about the appropriate diagnostic, therapeutic, and preventive strategies for VAP.
Multiple defense mechanisms protect the normal human respiratory tract from infection. Examples include anatomic barriers, such as the glottis and larynx; cough reflexes; constituents of tracheobronchial secretions; mucociliary clearance; epithelial lining fluid and surfactant components; cell-mediated and humoral immunity; a phagocytic system that involves alveolar macrophages and recruited neutrophils; and both humoral and cell-mediated adaptive immunity. When these coordinated components function properly, invading microbes are eliminated and clinical disease is avoided. When these defenses are impaired, or if they are overcome by a high inoculum of organisms or organisms of unusual virulence, pneumonitis results.
As suggested by the infrequent association of VAP with bacteremia, most of these infections appear to result from aspiration of potential pathogens that have colonized the mucosal surfaces of the oropharyngeal airways, the dental plaque, and/or the paranasal sinuses. Endotracheal intubation not only compromises the natural barrier between the oropharynx and trachea, but also facilitates the entry of bacteria into the lungs by pooling and leakage of contaminated secretions around the endotracheal tube cuff from the subglottic area to below the true vocal cords. Contaminated secretions leak into the lungs of most intubated patients and may be facilitated by the supine position. Moreover, biofilm formation on the inner and outer surfaces of the endotracheal tube provides a protected environment for pathogens. Bacterial aggregates in biofilm dislodged during suctioning are dangerous for the lung, because they are difficult to clear by host immune defenses and difficult to eradicate with antibiotics.
Tracheobronchial colonization with gram-negative bacilli (GNB) usually precedes the onset of VAP. Risk factors for tracheobronchial colonization appear to be the same as those that favor pneumonia and include advanced age, more severe illness, longer hospitalization, prior or concomitant use of antibiotics, malnutrition, endotracheal intubation, depressed level of consciousness, immune suppression from disease or medication, azotemia, underlying pulmonary disease, and longer duration of mechanical ventilation. Experimental studies have linked some of these risk factors with changes in adherence of GNB to respiratory epithelial cells. Although formerly attributed to loss of cell surface fibronectin, these changes in adherence also reflect alterations of cell surface carbohydrates. Bacterial adhesins and prior antimicrobial therapy appear to facilitate the process. Interestingly, Enterobacteriaceae usually appear first in the oropharynx, whereas Pseudomonas aeruginosa more often appear first in the trachea.
Although the stomach can be a reservoir for potential pneumonia pathogens, the gastropulmonary route of infection is not the primary route of infection in most critically ill patients. Progression of colonization from the stomach to the upper respiratory tract with subsequent episodes of VAP could not be demonstrated in several studies, and efforts to eliminate the gastric reservoir with antimicrobial therapy without decontaminating the oropharyngeal cavity have generally failed to prevent VAP. In fact, there is more than one potential pathway for colonization of the oropharynx and trachea in such a setting, including cross infection from the hands of health care personnel and contaminated respiratory therapy equipment. Patient care activities, such as bathing, oral care, tracheal suctioning, enteral feeding, and tube manipulations, provide many opportunities for transmission of pathogens when meticulous infection control practices are not followed.
The exact incidence of VAP varies widely depending on the case definition of pneumonia and the population being evaluated. For example, the incidence of VAP may be up to two times greater in patients when diagnosis is made by qualitative or semiquantitative sputum cultures rather than quantitative cultures of lower respiratory tract secretions. However, studies have confirmed that nosocomial pneumonia is considerably more frequent in ventilated patients than in other ICU patients, with the incidence being as much as 6-fold to 20-fold higher in ventilated patients than in nonventilated patients. VAP manifests in 9% to 27% of all intubated patients, and its incidence increases with duration of intubation. The risk of VAP is highest early in the course of the hospital stay—estimated to be 3% per day during the first 5 days of intubation, 2% per day during days 5 to 10 of intubation, and 1% per day after day 10. Because most mechanical ventilation is short term, approximately half of all episodes of VAP develop within the first 4 days of mechanical ventilation. In a large epidemiologic study, independent predictors of VAP determined by multivariate analysis were as follows: primary admitting diagnosis of burns, trauma, central nervous system disease, respiratory disease, or cardiac disease; mechanical ventilation during the preceding 24 hours; witnessed aspiration; and use of paralytic agents. Exposure to antibiotics conferred protection, but this effect was attenuated over time. According to four studies, the VAP rate was higher in patients with acute respiratory distress syndrome (ARDS) than in other ventilated patients, affecting between 34% and more than 70% of patients with ARDS and often leading to the development of sepsis, multiple organ failure, and death.
Attributable Mortality, Morbidity, and Costs
In mechanically ventilated patients in the ICU, those with VAP appear to have a 2-fold to 10-fold higher risk of death than those without pneumonia. Although these statistics indicate that VAP can be lethal, previous studies have not clearly demonstrated that pneumonia is responsible for the higher mortality rate of these patients. It is often difficult to determine whether ICU patients with severe underlying illness would have survived if they had not developed VAP. VAP, however, has been recognized in several case-control studies or studies using multivariate analysis as an important prognostic factor for different groups of critically ill patients. Based on a multistate progressive disability model that appropriately handled VAP as a time-dependent event in a high-quality database of 2873 mechanically ventilated patients, VAP-attributable mortality was found to be 8.1% overall. These results are consistent with those obtained in other observational studies using also multistate model and causal analysis, detecting a relatively limited VAP-attributable mortality.
Other factors beyond the simple development of VAP, such as the severity of the disease, the adequacy of antimicrobial therapy, or the responsible pathogens, may be more important determinants of outcome for patients in whom pneumonia develops. Indeed, it may be that VAP increases mortality only in the subset of patients with intermediate severity of illness, when initial treatment is inappropriate, and/or in patients with VAP caused by high-risk pathogens, such as P. aeruginosa. Patients with very low severity and early-onset pneumonia caused by organisms such as Haemophilus influenzae or Streptococcus pneumoniae have excellent prognoses with or without VAP, whereas very ill patients with late-onset VAP would be unlikely to survive.
Studies have shown clearly that patients with VAP have prolonged duration of mechanical ventilation and lengthened ICU and hospital stay than do patients who do not have VAP. Summarizing available data, VAP appears to extend the ICU stay by at least 4 days, with the attributable ICU length of stay being longer for medical than surgical patients and for patients infected with “high-risk” rather than “low-risk” organisms. The prolonged hospitalization of patients with VAP underscores the considerable financial burden imposed on the health care system by the development of VAP.
Microorganisms responsible for VAP differ according to the population of ICU patients, the duration of hospital and ICU stays, and the specific diagnostic methods used to establish the responsible pathogens. A number of studies have shown that GNB cause many of the respiratory infections in this setting. The data from 24 studies conducted on ventilated patients, for whom bacteriologic studies were restricted to uncontaminated specimens, confirmed these results: GNB represented 58% of recovered organisms ( Fig. 34-1 ). The predominant GNB were P. aeruginosa and Acinetobacter spp, followed by Proteus spp, Escherichia coli, Klebsiella spp, and H. influenzae . A relatively high rate of gram-positive pneumonias was also reported in those studies, with Staphylococcus aureus involved in 20% of the cases. Many episodes of VAP are caused by multiple pathogens.
Underlying diseases may predispose patients to infection with specific organisms. Patients with chronic obstructive pulmonary disease are at increased risk for H. influenzae , Moraxella catarrhalis, or S. pneumoniae infections; cystic fibrosis increases the risk for P. aeruginosa and/or S. aureus infections, while trauma and neurologic disease increase the risk for S. aureus infection. The causative agent for pneumonia also differs among ICU surgical populations, with 18% of the nosocomial pneumonias caused by Haemophilus spp or pneumococci, particularly in patients with trauma. Haemophilus spp and pneumococci are much less frequent causes of pneumonia in other surgical ICU patients, such as those with malignancy, organ transplants, or abdominal or cardiovascular surgery.
Despite somewhat different definitions of early-onset pneumonia, varying from onset of less than 3 to less than 7 days, high rates of infection with H. influenzae , S. pneumoniae , methicillin-sensitive S. aureus, or susceptible Enterobacteriaceae were consistently found in early-onset VAP, whereas P. aeruginosa, Acinetobacter spp, methicillin-resistant S. aureus (MRSA), and multiresistant GNB were significantly more frequent in late-onset VAP. The different pattern of distribution of etiologic agents between early- and late-onset VAP is linked to prior antimicrobial therapy in many patients with late-onset VAP. When multivariate analysis was used to identify risk factors for VAP caused by drug-resistant bacteria such as MRSA, P. aeruginosa, Acinetobacter baumannii , and/or Stenotrophomonas maltophilia in 135 consecutive episodes of VAP, only three variables remained significant: duration of mechanical ventilation of longer than 7 days before onset of VAP, prior antibiotic use, and prior use of broad-spectrum drugs (third-generation cephalosporins, fluoroquinolones, and/or imipenem). Not all studies have confirmed this distribution pattern, and in some studies the most common pathogens associated with early-onset VAP were P. aeruginosa , MRSA, and Enterobacter spp, with similar pathogens associated with late-onset VAP. These findings might be explained in part by prior hospitalization and the use of antibiotics before transfer to the ICU.
Legionella spp, anaerobes, fungi, viruses, and even Pneumocystis jirovecii are also potential causative agents, but these microbes are not commonly found when pneumonia is acquired during mechanical ventilation. Several of these causative agents, including viruses, might be more common than reported, because they are difficult to identify. Isolation of fungi, most frequently Candida species, at significant concentrations poses interpretative problems. Invasive disease has been reported in VAP, but yeasts are isolated more frequently from respiratory tract specimens in the absence of apparent disease. The use of the commonly available respiratory sampling methods (bronchoscopic or nonbronchoscopic) in ventilated patients is not sufficient to make the diagnosis of Candida pneumonia, and evidence of lung tissue invasion is also needed.
Two diagnostic strategies can be used when VAP is suspected, typically when a patient has new or progressive radiographic infiltrates and clinical findings suggesting infection, such as the new onset of fever, purulent sputum, leukocytosis, and a decline in arterial oxygenation. The first strategy is to treat every patient clinically suspected of having a pulmonary infection with new antibiotics, even when the likelihood of infection is low, arguing that several studies showed that immediate initiation of appropriate antibiotics was associated with reduced mortality. The second strategy is to use an invasive diagnostic approach based on quantitative cultures of distal respiratory specimens obtained using bronchoscopic or nonbronchoscopic techniques, such as bronchoalveolar lavage (BAL) or a protected specimen brush (PSB), in order to improve the identification of patients with true VAP and facilitate decisions about whether or not to treat with antibiotics. Although no consensus exists on the best diagnostic strategy for patients clinically suspected of having VAP, the goal of each strategy is to institute early appropriate antibiotic therapy in patients with true VAP and to withhold it in others.
The Clinical Diagnostic Strategy
With the clinical strategy, all patients suspected of having VAP are treated with new antibiotics. The selection of appropriate empirical therapy is based on risk factors and local microbiologic and resistance patterns, and involves qualitative testing to identify possible pathogens. The initial antimicrobial therapy is adjusted according to culture results or clinical response ( Fig. 34-2 ). Antimicrobial treatment is discontinued only if the following three criteria are fulfilled on day 3: (1) clinical diagnosis of VAP is unlikely (there are no definite infiltrates found on chest radiography at follow-up and no more than one of the three following findings is present: temperature greater than 38.3° C, leukocytosis or leukopenia, and purulent tracheobronchial secretions) or an alternative noninfectious diagnosis is confirmed; (2) tracheobronchial aspirate culture results are nonsignificant; and (3) severe sepsis or shock is not present.
This clinical approach has two undisputable advantages: first, no specialized microbiologic techniques are required, and, second, the risk of missing a patient who needs antimicrobial treatment is minimal when all suspected patients are treated with new antibiotics.
While the simple qualitative culture of endotracheal aspirates (EAs) is a technique with a high percentage of false-positive results due to bacterial colonization of the proximal airways in many ICU patients, studies using quantitative culture techniques suggest that the diagnostic accuracy of EA cultures is similar to the accuracy of more invasive techniques. The inherent advantages of these techniques are that they are less invasive, they are available to nonbronchoscopists, they are less expensive than bronchoscopy, they are less likely to compromise gas exchange, and they can be performed in patients with small endotracheal tubes. The disadvantages include the potential sampling errors inherent in a blind technique and lower specificity for distinguishing airway colonization from true pneumonia.
Another option when using the clinical approach is to follow the strategy described by Singh and colleagues, in which decisions regarding initial antibiotic therapy are based on a clinical score constructed from seven variables, the Clinical Pulmonary Infection Score (CPIS). Patients with a CPIS greater than 6 are considered to have VAP and are treated with antibiotics for 10 to 21 days; if the CPIS score is 6 or less, antibiotics are discontinued after 3 days ( Fig. 34-3 ). Such an approach avoids prolonged treatment of patients who have a low likelihood of infection, while allowing immediate treatment of patients who are more likely to have VAP. Two conditions must be fulfilled when using this strategy. First, the selection of initial antimicrobial therapy should be based on the most common microbes responsible for VAP at each institution. For example, ciprofloxacin would not be the right choice in hospitals with a high prevalence of MRSA infections. Second, physicians should reevaluate antimicrobial treatment on day 3, when susceptibility patterns of the microorganisms recovered from pulmonary secretions are available, in order to select treatment with a narrower spectrum antibiotic.
The Invasive Diagnostic Strategy
With the invasive strategy, quantitative cultures of lower respiratory secretions (BAL or PSB collected with or without a bronchoscope) are used to define both the presence of pneumonia and the etiologic pathogen. Growth above a threshold concentration is required to make a diagnosis of VAP and determine the causative microorganisms. Growth below the threshold is assumed to be due to colonization or contamination. Using this strategy, therapeutic decisions are made according to a strict protocol, using the results of direct examination of distal pulmonary samples and results of quantitative cultures in deciding whether to start antibiotic therapy, which pathogens are responsible for infection, which antimicrobial agents to use, and whether to continue therapy ( Fig. 34-4 ).
Quantitative cultures of BAL and/or PSB specimens consistently yield fewer microorganisms above the diagnostic threshold than are present in qualitative cultures of tracheal aspirates. Thus, when therapeutic decisions are based on these data, fewer patients are treated with antibiotics and a potentially narrower spectrum of therapy is used than when using the clinical approach, thereby limiting the emergence and dissemination of drug-resistant strains and minimizing antibiotic-related toxicity.
Another compelling argument in favor of the invasive strategy is that this approach directs attention away from the lungs as the source of fever when BAL/PSB quantitative culture results are negative. Many ICU patients with negative bronchoscopic cultures have other potential sites of infection, such as the wounds, the urinary tract, or intravascular catheters, that need to be identified in order to avoid delays in initiating appropriate treatment.
The accuracy of bronchoscopic techniques is questionable in patients who have received prior antibiotics, particularly when new antibiotics are introduced after the onset of the symptoms suggestive of nosocomial pneumonia and before pulmonary secretions are collected. When pneumonia develops in patients who have been receiving systemic antibiotics for several days, cultures of respiratory secretions are not modified in a major way as long as the samples for culture are obtained before initiating the new antibiotics, because the bacteria responsible for the new infection are likely to be resistant to the antibiotics that were used previously. However, if samples for microbiologic cultures of pulmonary secretions are obtained after initiation of new antibiotics in patients suspected of having VAP, the newly initiated antibiotics can increase the number of false-negative results, regardless of the way in which the secretions are obtained.
One major technical problem with all bronchoscopic techniques is proper selection of the sampling area in the tracheobronchial tree. The sampling area is usually selected based on the location of the radiographic infiltrate or on the bronchoscopic identification of a pulmonary segment that has purulent secretions. In patients with diffuse pulmonary infiltrates or minimal new changes in a previously abnormal chest radiograph, determining the correct segment to sample can be difficult. In such cases, sampling should be directed to the area where endobronchial abnormalities are maximal. Because autopsy studies indicate that VAP frequently involves the posterior portion of the right lower lobe, this area should probably be given priority for sampling.
Summary of the Evidence
Aside from decision-analysis studies and a single retrospective study, five trials to date have used a randomized scheme to assess the effect of a diagnostic strategy on antibiotic use and outcome in patients suspected of having VAP. In three randomized studies conducted in Spain, no differences were found in mortality and morbidity when either invasive (PSB and/or BAL) or noninvasive (quantitative endotracheal aspirate cultures) techniques were used to diagnose VAP. These studies were relatively small, ranging from 51 to 88 patients. Antibiotics were continued in all patients despite negative cultures, thereby offsetting the potential advantage of the specific diagnostic test in patients with suspected VAP. Several prospective studies have concluded that antibiotics can be stopped in patients with negative quantitative cultures, without adversely affecting the recurrence of pneumonia and mortality.
In a randomized French study of 413 patients, those managed with an invasive strategy using BAL and/or PSB had a lower mortality rate on day 14, lower sepsis-related organ failure assessment scores on day 3 and 7, and less antibiotic use. Of note, in the invasive strategy group, 22 nonpulmonary infections were diagnosed, whereas in the clinical strategy group, only five were diagnosed, suggesting that physicians using the clinical strategy overdiagnosed VAP and thereby failed to identify nonpulmonary infections. A randomized trial conducted by the Canadian Critical Care Trials Group investigated the effect of different diagnostic approaches on outcomes of 740 patients suspected of having VAP. There was no difference in the 28-day mortality rate in patients in whom BAL was used compared with those in whom endotracheal aspiration was used as the diagnostic strategy. The BAL group and the endotracheal aspiration group also had similar rates of targeted antibiotic therapy on day 6, days alive without antibiotics, and maximum organ dysfunction scores. Unfortunately, information about how the decision algorithms were followed in the two diagnostic arms once cultures were available was not provided, raising uncertainties about how de-escalation of antibiotic therapy was pursued in patients with negative BAL cultures. The potential benefit of using a diagnostic tool such as BAL for safely restricting unnecessary antimicrobial therapy in such a setting can be obtained only when decisions regarding antibiotics are closely linked to bacteriologic results, including both direct examination and cultures of respiratory specimens.
Antimicrobial therapy of patients with VAP is a two-stage process. The first stage involves administering broad-spectrum antibiotics to avoid inadequate treatment of patients with true bacterial pneumonia. The second stage focuses on trying to achieve this objective without overusing or abusing antibiotics. In general, the first goal can be accomplished by identifying patients with pneumonia in a rapid fashion and starting therapy with an empirical regimen that is likely to treat the most common etiologic agents in a particular institution. This requires that the initial antibiotic choice be driven by knowledge of the likely etiologic pathogens, and the local patterns of antimicrobial resistance. The second goal involves stopping therapy in patients with a low probability of VAP, focusing and narrowing treatment once the etiologic agent is known, switching to monotherapy after day 3 whenever possible, and shortening the duration of therapy to 7 to 8 days in most patients, as determined by the patient’s clinical response and information about the bacteriology ( Table 34-1 ).
|Step 1: Start therapy using broad-spectrum antibiotics||Due to the emergence of multiresistant GNB, such as P. aeruginosa and ESBL-producing GNB in many institutions, and the increasing role of MRSA, empirical treatment with broad-spectrum antibiotics is justified in most patients with a clinical suspicion of VAP.|
|Step 2: Stop therapy if the diagnosis of infection becomes unlikely||The goal is to ensure that ICU patients with true bacterial infection receive immediate appropriate treatment. However, this can result in more patients receiving antimicrobial therapy than necessary, because clinical signs of infection are nonspecific.|
|Step 3: Use narrower spectrum antibiotics once the etiologic agent is identified||For many patients with VAP, including those with late-onset infection, therapy can be narrowed once the results of respiratory tract and blood cultures are available, either because an anticipated organism (e.g., P. aeruginosa and Acinetobacter spp. or MRSA) was not recovered, or because the organism isolated is sensitive to a more narrow-spectrum antibiotic than used in the initial regimen. *|
|Step 4: Use pharmacokinetic-pharmacodynamic data to optimize treatment||Clinical and bacteriologic outcomes can be improved by optimizing the therapeutic regimen according to pharmacokinetic and pharmacodynamic properties of the agents selected for treatment.|
|Step 5: Switch to monotherapy on days 3 to 5||There are no clinical benefits to using a regimen combining two antibiotics for more than days 3 to 5, provided that initial therapy was appropriate, the clinical course appears favorable, and microbiologic data do not point to a very difficult-to-treat microorganism.|
|Step 6: Shorten the duration of therapy||Reducing duration of therapy in patients with VAP has led to good outcomes with less antibiotic use. Prolonged therapy leads to colonization with antibiotic-resistant bacteria, which may precede a recurrent episode of VAP.|
* Chastre J, Fagon JY: Pneumonia in the ventilator-dependent patient. In Tobin MJ, editor: Principles and practice of mechanical ventilation , New York, 1994, McGraw-Hill, pp 857–890; and Rello J, Vidaur L, Sandiumenge A, et al: De-escalation therapy in ventilator-associated pneumonia. Crit Care Med 32:2183–2190, 2004.
Failure to initiate prompt appropriate therapy (i.e., using an agent to which the etiologic organism is sensitive, with the optimal dose and route of administration) has been consistently linked with increased mortality in patients with VAP. Due to the emergence of multiresistant GNB, such as P. aeruginosa , extended-spectrum β-lactamase-producing Enterobacteriaceae, and carbapenemase-producing Klebsiella pneumoniae, and the increasing role of gram-positive bacteria, such as MRSA, empirical treatment with broad-spectrum antibiotics is justified in most patients with a clinical suspicion of VAP. The choice of agents should be based on local patterns of antimicrobial susceptibility and anticipated side effects, and should take into account the antibiotics that the patients have received within the prior 2 weeks, striving not to use the same antimicrobial classes, if possible. Having current knowledge about local bacteriologic patterns can increase the likelihood that appropriate initial antibiotic treatment will be prescribed. Only patients with early-onset infection and no specific risk factors, such as prolonged duration of hospitalization, admission from a health care–related facility, and recent prolonged antibiotic therapy, can be treated with a relatively narrow-spectrum drug, such as a nonpseudomonal third-generation cephalosporin.
Several published reports have demonstrated the need to adjust the target dose of antimicrobial agents used in treating severe VAP to the individual patient’s pharmacokinetics and the putative bacterial pathogens’ susceptibilities. Most investigators distinguish between antimicrobial agents that kill by a concentration-dependent mechanism (e.g., aminoglycosides and fluoroquinolones) from those that kill by a time-dependent mechanism (e.g., β-lactams and vancomycin). Altered pharmacokinetics secondary to an increase in the volume of distribution in critically ill patients can result in insufficient serum β-lactam concentrations when standard doses are administered, emphasizing the need to monitor peak and trough levels of antibiotics carefully when treating resistant pathogens. Higher dosing regimens than those usually recommended and/or prolonged duration of infusion are frequently needed in such circumstances. Development of a priori dosing algorithms based on minimal inhibitory concentrations, patient creatinine clearance and weight, and the clinician-specified area under the inhibitory curve target might be a valid way to improve treatment of these patients, leading to a more precise approach than current guidelines for use of antimicrobial agents.
Avoiding the Overuse of Antibiotics
The need to ensure that ICU patients with true bacterial infections promptly receive an appropriate antibiotic regimen can lead to many more patients receiving antimicrobial therapy than is actually necessary, because clinical signs of infection are relatively nonspecific. Thus, when a clinical approach to VAP is used, it is important to perform serial clinical and microbiologic evaluations, and to reevaluate therapy after 48 to 72 hours so that it can be stopped if infection is unlikely. To accomplish this, all diagnostic strategies that are designed for managing patients with a clinical suspicion of VAP should make explicit the decision tree that is used to identify patients with a low probability of infection, in whom therapy can be stopped when infection appears improbable.
For many patients with VAP, including those with late-onset infection, therapy can be narrowed once the results of respiratory tract and blood cultures are available, if no resistant organism (e.g., P. aeruginosa , Acinetobacter spp or MRSA) is recovered or if the isolated organism is sensitive to a narrower spectrum antibiotic. For example, vancomycin and linezolid should be stopped if MRSA is not identified, unless the patient is allergic to β-lactams or has developed an infection caused by a gram-positive microorganism. Very-broad-spectrum agents, such as carbapenems, piperacillin-tazobactam, and/or cefepime, should also be restricted to patients with infection caused by pathogens that are only susceptible to these agents. Clinicians must be aware that the emergence of resistant variants may lead to treatment failure when third-generation cephalosporins are chosen to treat infections caused by Enterobacter, Citrobacter, Morganella morganii , indole-positive Proteus , and Serratia spp, due to the presence of inducible β-lactamases, even if the isolate is initially characterized as susceptible.
The most common reason to use combination therapy in initial management of patients with VAP is to achieve synergy in treating P. aeruginosa or other difficult-to-treat GNB. However, antibiotic synergy has been shown to be valuable only in vitro and in patients with neutropenia or bacteremic infection, which is uncommon in VAP. A recent meta-analysis evaluated all prospective randomized trials of β-lactam monotherapy compared with β-lactam/aminoglycoside combination regimens in 7586 patients with sepsis, of whom at least 1200 patients had VAP. The clinical success rates were similar with monotherapy versus combination therapy, and combination therapy had no advantage in the treatment of P. aeruginosa infections. Importantly, combination therapy did not prevent the emergence of antimicrobial resistance during treatment, but it was associated with a significantly higher rate of nephrotoxicity. Based on these data, therapy can be switched to monotherapy in most patients after 3 or 5 days, as long as initial therapy is appropriate, the clinical course is favorable, and microbiologic data do not identify a difficult-to-treat microorganism with a high in vitro minimal inhibitory concentration, as is found with some lactose-nonfermenting GNB.
Efforts to reduce the duration of therapy for VAP are justified by studies of the natural history of the response to therapy. Most patients with VAP who receive appropriate antimicrobial therapy have a good clinical response within the first 6 days. Prolonged therapy promotes colonization with antibiotic-resistant bacteria, which may lead to a recurrent episode of VAP. A multicenter randomized controlled trial of 401 patients with microbiologically proven VAP showed that the clinical outcomes of patients who received appropriate empirical therapy for 8 days were similar to those of patients who received therapy for 15 days. A trend to greater rates of relapse for short-duration therapy was seen when the etiologic agent was P. aeruginosa or Acinetobacter spp, but the clinical outcomes were indistinguishable. These results were confirmed in two later studies, including a prospective randomized trial of 290 patients evaluating an antibiotic discontinuation policy. Possible exceptions to this recommendation include immunosuppressed patients, those whose initial antimicrobial treatment was not appropriate for the causative microorganisms, and patients whose infection was caused by nonfermenting GNB and had no improvement in clinical signs of infection.
Many clinicians remain hesitant about prescribing antibiotics for fewer days for patients with VAP, and they prefer to customize antibiotic duration based on the clinical course of the disease and/or using serial determinations of a biomarker such as procalcitonin . The rationale for using a biomarker to tailor antibiotic treatment duration relies on evidence that the inflammatory response is often proportional to infection severity. When that response is absent or low, it might be logical to discontinue antibiotics earlier. Thus, adapting antimicrobial treatment duration to procalcitonin kinetics seems reasonable and has been demonstrated as useful in several randomized trials targeting patients with acute respiratory infection, including five trials conducted in the ICU.
Because insufficient delivery of antibiotics to the site of infection in patients with VAP may lead to clinical and microbiologic failures, efforts to increase pulmonary delivery of antimicrobial agents have been investigated. Delivering drugs via aerosolization is one approach, assuming that this technique actually promotes higher drug concentrations at the infected site. By achieving high pulmonary antibiotic concentrations, this mode of administration could increase the antibacterial activity of concentration-dependent antibiotics, such as aminoglycosides, or provide bactericidal activity of antibiotics in infections caused by pathogens of impaired sensitivity. By limiting systemic exposure, it could also allow the administration of antibiotics with high systemic toxicity, such as aminoglycosides and polymyxins.
Several studies, based on nebulizers with improved technology, have renewed the interest in aerosolized antibiotic therapy for VAP. In anesthetized piglets on mechanical ventilation for severe E. coli bronchopneumonia, amikacin lung tissue concentrations were markedly higher following aerosolization as compared to intravenous administration. In a study using a device with a vibrating plate and multiple apertures to produce an aerosol of amikacin, the nebulized drug was well-distributed in the lung parenchyma, with high tracheal and alveolar levels and serum concentrations below the renal toxicity threshold. Aerosolized amikacin was well-tolerated, without any severe adverse event, and patients who received amikacin twice daily required significantly fewer other antibiotics than patients given placebo.
Aerosolized polymyxin is also being used to treat infections caused by multidrug-resistant GNB, mainly Acinetobacter baumannii, P. aeruginosa, and carbapenemase-producing K. pneumoniae , with mixed results. In a randomized trial of 100 patients with VAP due to GNB (predominantly multi-drug resistant A. baumannii and/or P. aeruginosa ), patients treated with a combination of systemic antibiotics and nebulized colistin had a higher rate of favorable microbiologic outcome compared with patients treated with systemic antibiotics alone (microbiologic eradication or presumed eradication 61% vs. 38%), but there were no differences in the rate of favorable clinical outcomes (51% vs. 53%). In a retrospective case-control study of 86 patients with VAP due to multidrug-resistant GNB (predominantly A. baumannii ) treated with a combination of intravenous and aerosolized colistin compared with intravenous colistin alone, there was only a trend toward improved rates of clinical cure, pathogen eradication, and mortality in the patients who received aerosolized and intravenous colistin.
Thus, although recent investigations emphasize the potential contribution of aerosolized antibiotics to treat VAP as an adjunctive therapy to intravenous antibiotics, the clinical impact of such a strategy has not been established. At present, aerosolized antibiotics can only be recommended to treat patients with multidrug-resistant VAP for which no effective intravenous antibiotics are available. Large prospective trials are needed to evaluate the potential usefulness of this therapeutic modality.
Because VAP is associated with increased morbidity, longer hospital stay, increased health care costs, and higher mortality rates, prevention is an important goal.
Conventional Infection Control Approaches
The design of the ICU has a direct effect on the potential for nosocomial infections. Adequate space and lighting, properly functioning ventilation systems, and appropriate handwashing facilities all lead to lower infection rates. It is important to note, however, that upgrading the physical environment does not reduce the infection rate unless the attitudes and practices of health care personnel are also improved. In any ICU, one of the most important factors is the health care staff, including the number, quality, and motivation of medical, nursing, and ancillary members. The team should include a sufficient number of nurses to minimize their movement from one patient to another and to avoid having them work under constant pressure. The importance of personal cleanliness and attention to aseptic procedures must be emphasized at every opportunity. It is clear that careful monitoring, decontamination, and compliance with guidelines for the use of respiratory equipment all reduce the incidence of nosocomial pneumonia. Handwashing and hand rubbing with alcohol-based solutions remain the most important components of effective infection control practices in the ICU.
Environmental and patient-oriented microbiologic monitoring facilitates the early recognition of colonization and infection, and has been associated with significant reductions in nosocomial infection rates. The focal point for infection control activities in the ICU is a surveillance system designed to establish and maintain a database that identifies endemic rates of nosocomial infections. This information facilitates the recognition of epidemics, when infection rates rise above the endemic threshold for a specific type of nosocomial infection.
An antibiotic policy that restricts the prescription of broad-spectrum agents and inappropriate antibiotics is of major importance. Better use of antibiotics in the ICU can be achieved by implementing strict guidelines, avoiding the treatment of patients who do not have bacterial infections, using narrow-spectrum antibiotics whenever possible, and reducing the duration of treatment. Similarly, transfusion of red blood cells and other allogenic blood products should follow a strict policy, because several studies have identified exposure to allogenic blood products as a risk factor for postoperative infection and pneumonia.
Specific Prophylaxis Against Ventilator-Associated Pneumonia
Specific strategies aimed at reducing the duration of mechanical ventilation (a major risk factor for VAP), such as improved methods of sedation, use of protocols to facilitate and accelerate weaning, using adequate levels of positive end-expiratory pressure, and using intensive insulin therapy to control blood glucose are integral parts of any infection control program. All are based on the application of strict protocols. Similarly, noninvasive positive-pressure ventilation using a face mask should be employed whenever possible.
Some very simple, safe, inexpensive, and logical measures may have major effects on the frequency of VAP in mechanically ventilated patients. These include avoiding nasal insertion of endotracheal and gastric tubes, maintaining the endotracheal tube cuff pressure above 20 cm H 2 O to prevent leakage of bacteria around the cuff into the lower respiratory tract, promptly reintubating patients who are likely to fail extubation, keeping patients in the semirecumbent position, especially when enteral nutrition is used, removing tubing condensate, and providing adequate oral hygiene with an antiseptic such as chlorhexidine.
Continuous or intermittent suctioning of oropharyngeal secretions has been proposed as a means to avoid chronic aspiration of secretions through the tracheal cuff of intubated patients. Aspiration of subglottic secretions requires the use of a specially designed endotracheal tube with a separate lumen that opens into the subglottic region. Thirteen randomized controlled trials with a total of 2442 randomized patients have studied aspiration of subglottic secretions for the prevention of VAP. Of the 13 studies, 12 reported a reduction in VAP rates in the subglottic secretion drainage arm. When the results were combined in a meta-analysis, the overall risk ratio for VAP was 0.55 (95% CI, 0.46-0.66; P < 0.00001) with no heterogeneity, and the use of subglottic secretion drainage was associated with reduced ICU length of stay, decreased duration of mechanical ventilation, and increased time to first episode of VAP. However, there was no effect on hospital or ICU mortality. Preliminary data in animal models and from small randomized human studies support the hypothesis that an endotracheal tube coated externally and internally with a potent antiseptic product such as silver could have a sustained antimicrobial effect within the proximal airways and block biofilm formation at its surface. Such a device was evaluated in a large, randomized, multicenter, single-blind trial. The authors conclude that the new device was able to lower the VAP frequency from 7.5% for the control group to 4.8% for the group receiving the silver-coated endotracheal tube. The silver-coated tube, however, did not reduce mortality rates, the duration of intubation, hospital length of stay, or the frequency or severity of adverse effects.
Gastric colonization by potentially pathogenic organisms increases with decreasing gastric acidity. Thus, medications that decrease gastric acidity (antacids, histamine 2 [H 2 ] blockers) can increase the gastric bacterial burden and increase the risk of VAP; medications that do not affect gastric acidity (e.g., sucralfate) do not appear to increase this risk. Several meta-analyses of more than 20 randomized trials have evaluated the risk for VAP associated with methods used to prevent gastrointestinal bleeding in critically ill patients. The largest randomized trial comparing ranitidine to sucralfate showed that ranitidine was superior in preventing gastrointestinal bleeding and did not increase the risk of VAP. Therefore, despite the potential advantage of sucralfate (potentially less VAP with more gastrointestinal bleeding) over H 2 blockers (potentially more VAP with less gastrointestinal bleeding) in preventing VAP, stress ulcer prophylaxis with H 2 blockers appears to be safe in patients who are at high risk for bleeding as well as VAP. Although proton-pump inhibitors are now widely used for gastric bleeding prophylaxis in the ICU, based on their potentially higher efficacy, their use is associated with similar rates of nosocomial pneumonia as H 2 blockers.
Selective decontamination of the digestive tract (SDD) includes a short course of systemic antibiotic therapy, such as cefotaxime, trimethoprim or a fluoroquinolone, and topical administration of nonabsorbable antibiotics (usually an aminoglycoside, polymyxin B, and amphotericin) to the mouth and stomach, in order to eradicate potentially pathogenic bacteria and yeast that may cause infections. Since the original study published by Stoutenbeek and coworkers in 1984, which demonstrated a decrease of the overall infection rate in patients receiving the SDD regimen, more than 40 randomized controlled trials and eight meta-analyses have been published. All eight meta-analyses reported a significant reduction in the risk of VAP, and four reported a significant reduction in mortality. Recently, three prospective, randomized, controlled trials, all performed in ICUs with low rates of antibiotic resistance, have been published that were large enough to show a significant survival benefit in SDD-treated patients. All three were in favor of treatment with SDD, the largest and most recent one demonstrating a relative decrease in 28-day mortality rate (OR 0.83, 95% CI, 0.72 to 0.97) and an absolute survival benefit of 3.5%. Even so, widespread use of SDD in ICU patients remains controversial. The major concern with use of SDD is that it can promote the emergence of resistant bacteria, particularly gram-positive bacteria such as MRSA. This is likely to be even a greater problem in ICUs with a high baseline rate of resistance. In contrast to what was expected, however, most studies that have evaluated this issue showed a lower incidence of colonization with resistant bacteria in SDD-treated patients than in control patients. Putative explanations why colonization with resistant microorganisms is lower after treatment with SDD include the susceptibility of gram-negative aerobic bacteria to the commonly used combination of polymyxin E and tobramycin, the fact that treatment with polymyxin E rarely induces resistance, the very high local concentrations in the bowel of the used antibiotics, and the lower rate of use of systemic antibiotics in SDD-treated patients.
Implementing a Structured Prevention Policy
The application of consistent evidence-based interventions to prevent VAP has been highly variable from one ICU to another and is often suboptimal. Furthermore, no single preventive measure can succeed alone, emphasizing the need to use multifaceted and multidisciplinary programs to prevent VAP. Such programs are frequently referred to as “care bundles.” A care bundle is a set of readily implementable interventions that are required to be undertaken for each patient on a regular basis. The key goal is for every intervention to be implemented for every patient on every day of his or her stay in the ICU. Compliance is assessed for the bundle as a whole, so failure to complete even a single intervention means failure of the whole bundle at a particular assessment. The interventions need to be packaged in such a way that compliance is readily assessed, which usually means that no more than five interventions are included in each care bundle. The performance goal is to routinely achieve over 95% compliance. Care bundles make it possible to introduce evidence-based preventive measures, including appropriate nurse staffing levels, hand hygiene with alcohol-based formulations, standardized weaning protocols and daily interruption of sedation, oral care with chlorhexidine, and keeping patients who receive enteral nutrition in a semirecumbent position. Several studies using quasiexperimental design have confirmed the usefulness of this strategy for preventing VAP in the ICU.
The lack of methodologic rigor of the reported studies, however, precludes any conclusive statements about “bundle care” effectiveness or cost-effectiveness. The exact set of key interventions that should be part of the “VAP-prevention bundle” is also not currently known, nor are factors contributing to its success. Successful VAP prevention requires an interdisciplinary team, educational interventions, system innovations, process indicator evaluation, and feedback to health care workers. As shown in a recent study, simply having a checklist available for reference without consideration of a robust implementation and adherence strategy is unlikely to maximize patient outcomes.
In the United States, the Centers for Medicare and Medicaid Services has proposed stopping hospital reimbursements for care made necessary by preventable complications, including nosocomial infections, aiming for a zero-VAP rate. Although this plan may have the desirable consequences of improving the quality of care, it also may penalize hospitals that admit high-risk patients and inadvertently encourage institutions to underreport VAP or to overuse antibiotics, thereby favoring dissemination of multidrug-resistant microorganisms. This possibility further underscores the need to evaluate all new strategies potentially aimed at preventing VAP against current best clinical practices.