The United States Centers for Disease Control (CDC) published an updated surveillance framework in 2013 about ventilator-associated events (VAE). The rationale was to improve the objectivity of previous (2005) surveillance parameters for ventilator-associated complications. The framework was intended solely for surveillance purposes and not for clinical diagnosis and management.
Numerous pros and cons of the CDC surveillance framework (2013) have been outlined. Among the pros are greater objectivity of parameters; ease of electronic computation; ability to use surveillance parameters as a metric of quality of healthcare for ventilated patients; close association of the parameters to mortality rates and, an association with greater potential for multi-disciplinary collaboration in mitigation of VAE. These are based on findings from the surveillance framework. Limitations of the framework, however, include the fact that some VAEs may reflect the natural history of the critical illness rather than preventable, ventilator-induced complications; the low positive predictive value of VAE surveillance criteria for VAP and, the potential to miss cases of VAP. Additionally, VAP parameters were arguably non-specific to distinguish VAP from other forms of (potentially hospital-acquired) respiratory illness in critically ill patients.
The CDC (2013) has categorized VAE as (I) ventilator-associated conditions (VAC), (II) infection-related ventilator-associated complications (IVAC), and (III) possible ventilator-associated pneumonia (PVAP) or probable ventilator-associated pneumonia VAP.
Ventilator-associated Conditions (VAC)
VAC is defined as an increase in oxygen requirement, sustained for a minimum of 2 days in a mechanically ventilated patient. Day 1, in the context of VAE, is the day of intubation and commencement of mechanical ventilation. Sustained, increased oxygen requirement constitutes an increase in the daily, minimum baseline positive end-expiratory pressure (PEEP), equal to or greater than 3cm H2O or, an increase in the daily minimum fraction of inspired oxygen (FiO2) that is equal to or greater than 20 points for a minimum of 2 days. VAC may be defined only in those patients who have been mechanically ventilated for a minimum of 2 days and who have displayed pre-existing, stable, or improved oxygenation parameters before the period of increased oxygen requirement. Pre-existing improvement or stability is defined as two or more days of stable or decreasing FiO2 or PEEP, while on the mechanical ventilator. Baseline FiO2 and PEEP would be defined as the minimum daily FiO2 and PEEP measured during the period of improvement or stability.
Infection-related Ventilator-associated Complications (IVAC)
IVAC denotes clinical signs of infection in a patient with VAC. However, for the assessment of IVAC, the minimum duration of mechanical ventilation is three days, and patients must have displayed signs of infection in the two days preceding- or within two days after the assessment of VAC. The assessment of IVAC suggests infection as a cause of VAC. Mandatory clinical signs for an assessment of IVAC are low-grade pyrexia (>38C) or hypothermia (temperature <36C); leukocytosis (>12000 cells/mm3) or leukopenia (less than or equal to 4000 cells/mm3) and, there must have been the initiation of an antimicrobial agent with continuation for more than or equal to 4 days.
Possible entilator-associated Pneumonia (PVAP) and Probable Ventilator-associated Pneumonia
An assessment of VAC and IVAC should prompt suspicion of VAP. PVAP and probable VAP are respectively defined by the presence of IVAC and laboratory or microbiologic evidence of infection. The criteria for an assessment of PVAP include the presence of greater than or equal to 25 neutrophils and less than or equal to 10 squamous epithelial cells on sputum microscopy or a positive qualitative, semi-quantitative or quantitative culture on a specimen from the trachea, bronchi or lungs. Probable VAP is assessed based on a positive culture from endotracheal aspiration (greater than or equal to 10 to the power 5, colony-forming units [CFU]/mL) or a positive culture from bronchoalveolar lavage [BAL] (greater than or equal to 10 to the power 4, CFU/mL) or positive culture from protected brush sampling (greater than or equal to 10 to the power of 3, CFU/mL). These indicators may be superseded by a positive pleural fluid culture obtained on thoracocentesis, positive lung histopathology or, a positive culture of specific respiratory viruses (for example, respiratory syncytial virus) or Legionella, demonstrated in respiratory secretions.
In a retrospective cohort study conducted at a tertiary, academic center in 2013, among 20 356 mechanically ventilated patients, 5.6% (n=1141) of patients had VAC, 2.1% (n=431) had IVAC, 0.7% (n=139) had PVAP and 0.6% (n=127) had probable VAP. The risk of VAC was noted to be highest among medical, surgical, and thoracic units. Cardiac and neuroscience units displayed the lowest risk. Forty-two percent (42%) of cases of VAC and IVAC were from surgical units, and 29% were from medical units. Among cases of IVAC and probable VAP, Staphylococcus aureus (29%), Pseudomonas aeruginosa (14%), and Enterobacter species (7.9%) were the most commonly implicated organisms. Overall, VAE was associated with a greater number of days to extubation, a longer length of stay in the hospital, and a higher risk of mortality.
These findings were similar to those of a single-center, retrospective cohort study conducted at the general ICU of an academic hospital in 2017 in Tokyo, Japan. Among 407 adult patients who were mechanically ventilated for at least four days, higher mortality was noted among patients with VAC and IVAC (but no VAP), compared to those patients without VAE or VAP. VAC and IVAC, respectively, were shown to be independently associated with higher mortality.
A multi-state survey conducted in the United States in 2014 revealed that hospital-acquired pneumonia (HAP) and VAP are 2 of the most common types of hospital-acquired infections (HAIs) and comprise 22% of all HAI cases.
A recent meta-analysis (2013) evaluated evidence from studies conducted only in the United States between 1986 and 2013. VAP was found to be the second leading contributor (31.6%) to the total annual costs for five major HAIs ($9.8 billion [95% CI, $8.3-$11.5 billion]), after surgical site infections which contributed 33.7% of the total figure.
There is a need for further research on the epidemiology of VAE globally and among groups of low-to-middle income countries and higher-income countries, respectively.
Issues of Concern
Critically ill patients typically require endo- or nasotracheal intubation for mechanical ventilation. Intubation, as a procedure, has an inherent risk of complications. Among these are failure to intubate with a consequent requirement for repeat attempts; difficult intubation; laryngospasm, bronchospasm and airway obstruction; tube obstruction or occlusion; incorrect tube placement (right mainstem bronchial- or esophageal intubation); tube dislodgement; aspiration; severe hypoxia; severe hypotension; local trauma to the oral cavity, dentition, pharynx, larynx, and trachea; hematoma formation; tracheal stenosis or necrosis; upper respiratory tract infection including sinusitis; tracheobronchitis and, VAP. Multiple attempts at intubation tend to be associated with a greater risk of complications.
Ventilator-induced Lung Injury (VILI)
Atelectrauma, barotrauma, volutrauma, and biotrauma are four primary pathophysiologic mechanisms, which underlie VILI.
Atelectrauma is caused by high-shear forces that open and close recruitable atelectatic lung units. Shear stress and resultant mechanical damage develop at the interface of air boluses and collapsed recruited airways.
Atelectrauma is exacerbated by lung inhomogeneity. Alveoli are mechanically interdependent due to the presence of interalveolar septae. In the case of pathology, interalveolar septae between aerated alveoli and adjacent atelectatic or fluid-filled, non-aerated alveoli mediate atelectrauma. The collapse of or accumulation of fluid in one alveolus inadvertently causes deformation of neighboring alveoli, as the interalveolar septum deviates toward the collapsed or fluid-filled alveolus. This results in non-uniform inflation and abnormal shearing forces within neighboring alveoli during mechanical ventilation. The implications of lung inhomogeneity are, therefore, varying regional lung mechanics and a reduced overall lung volume available for optimal ventilation. On computed tomography (CT) of the chest, this may be visible as heterogeneous regions of well-aerated lung adjacent to areas of atelectasis or opacity. Histologically, regions of alveolar hyaline membranes, edema, and sloughing of the respiratory epithelium may be noted. Conditions characterized by lung inhomogeneity include atelectasis, ARDS, surfactant deficiency, and pulmonary edema.
Each inspiration requires pressure to: (I) overcome airway resistance, (II) accelerate air entry and, (III) overcome the effect of lung elasticity.
At the end of each inspiration, there is a transient lack of airflow in the lungs before expiration commences. The main distending pressure of the lung during this period is the transpulmonary pressure (i.e., alveolar airway pressure or plateau pressure minus pleural pressure), which has an inversely proportional relationship to lung volume.
Alveolar pressure may be easily estimated; it is the airway pressure at the end of inspiration during a period of no airflow. Similarly, the plateau pressure constitutes the airway pressure at the end of inspiration required to distend the lungs and chest wall, during a period of no airflow in a mechanically ventilated patient with no spontaneous breathing efforts. Pleural pressure is difficult to measure; it may be estimated by measurement of esophageal pressure, which is generally cumbersome and yields imprecise results. The plateau pressure is, therefore, commonly used in clinical practice as a stand-in, representative indicator (in place of transpulmonary pressure) of lung over-distension. In a patient with a pathologically stiff chest wall, much of the pressure exerted by the ventilator may be used to distend the chest wall, rather than the lungs. Therefore, a high plateau pressure does not necessarily indicate a high pulmonary distension force (i.e., high transpulmonary pressure). It is important to consider the influence of patients' underlying pathology when interpreting plateau pressure measurements.
In mechanically ventilated patients, barotrauma is caused by high lung inflation pressure, which in turn leads to high transpulmonary pressure, regional lung over-distension, and air leakage. Careful control of inflation pressure, therefore, serves as a strategy to limit lung distension and prevent barotrauma. Ventilation at high inflation pressures may cause an alveolar rupture, pneumothorax, pneumomediastinum, and subcutaneous emphysema, as forms of barotrauma or air leakage due to over-distension. Extremely low to negative pleural pressure may contribute to barotrauma at low airway pressures.
Volutrauma results from alveolar over-distension. Mechanical hyperinflation causes alveolar epithelial strain, which initiates mobilization of lipids to the alveolar plasma membrane for cell repair. The surface area of cells effectively increases, and the rupture of the plasma membrane is prevented. These protective mechanisms may become progressively overwhelmed, leading to cellular detachment from the plasma membrane under conditions of increasing cell strain. The junction between vascular endothelial cells and alveolar epithelium breaks down with resultant alveolar and interstitial edema.
Mechanical injury to the lungs may prompt an adverse inflammatory response, which may exert damaging effects, known as "biotrauma". Activation of injurious cytokines and other inflammatory mediators cause biotrauma not only in pathological and normal lung regions but also in other organs, with resultant multi-organ dysfunction and increased mortality. The respiratory epithelium in the lungs has a high surface area. Additionally, a substantial volume of blood circulation passes through the lungs per minute. The implication is that relatively small-scale, local inflammatory responses may precipitate a large release of pro-inflammatory cytokines with high potential for hematogenous spread and multi-organ damage. Concomitant physiologic impairment, for example, from sepsis, trauma, surgery, or chronic illness, predisposes patients to VILI from a cascading immune response.
"Baby lung" of Acute Respiratory Distress Syndrome (ARDS)
Patients with ARDS typically have impaired surfactant function and pulmonary edema. These factors contribute to the atelectasis of dependent lung regions. The total lung volume available for gaseous exchange is thereby reduced. This phenomenon is referred to as the ARDS "baby lung." Atelectasis of dependent lung regions is non-fixed and redistributable to other lung regions with a change of positioning; for example, anterior lung regions are more affected in prone positioning. Low tidal volumes may be required in ARDS because of reduced overall, functional lung volume, and the need to prevent regional over-distension. The recruitable lung regions during mechanical ventilation are not necessarily normal; other pathology of structure and function may be prevalent.
Therapeutic and Preventative Strategies for VILI
Therapeutic objectives in the context of mechanical ventilation have shifted from exclusive maintenance of gaseous exchange for survival, with minimal work of breathing to also include prevention of VILI. Accordingly, ventilator settings require adjustment in a manner that balances clinical benefits with potential risks.
Among preventative strategies are the use of lower tidal volumes (to prevent regional lung over-distension), the use of higher PEEP (to prevent atelectrauma), and delivery of sustained airway pressure of more than 35cm HO (to recruit atelectatic lung units).
Overall, there is no established, ideal ventilation strategy. Each intervention may also have inherent risks to physiologic function in addition to potential clinical benefits. For example, to mitigate the effect of auto-PEEP, lower tidal volumes may be necessary; however, there is a risk of a consequent rise in the arterial partial pressure of CO2 and the potential for respiratory acidosis and intracranial hypertension. Careful consideration of individual patient requirements, comorbidities, and risk factors is necessary as part of a holistic approach to the prevention of complications.
Low Tidal Volume
Critically ill patients tend to have less aerated, dependent lung regions, resulting in a reduced available lung volume for optimal ventilation. As a general principle, a smaller tidal volume should be delivered to normally aerated, non-dependent lung regions to prevent over-distension and barotrauma.
Alveolar collapse occurs commonly in respiratory failure. Low PEEP may be inadequate to aerate collapsed alveoli. Therefore, prolonged delivery of low PEEP may lead to atelectrauma. Contrarily, excessively high PEEP is associated with the risk of reduced venous return and cardiac output, as well as regional lung over-distension and barotrauma. It is thus necessary to maintain a level of PEEP, which balances the benefit of optimal recruitment with the risks of barotrauma and hemodynamic instability.
The measurement of transpulmonary pressure to guide PEEP delivery has shown benefit in improving oxygenation and lowering mortality rates; however, this is not routine in standard practice. It remains at the clinical discretion of attending physicians to institute recruitment maneuvers to optimize benefits and minimize risks of VILI and other complications until more standardized approaches to the achievement of this aim are ascertained.
High-frequency Oscillatory Ventilation (HFOV)
HFOV has been proposed as an intervention to minimize the risk of VILI. It involves the delivery of very small tidal volumes at high frequencies. Whether this intervention is associated with significantly improved clinical outcomes is, however, uncertain. Hence, the intervention does not form a part of standard practice.
Adjunctive Strategies to Prevent VILI
Reduction of Metabolic Demand
Limitation of patients' metabolic demands has been outlined as a therapeutic approach to prevent VILI. The rationale is that patient requirements for gaseous exchange may be decreased by decreasing metabolic activity. This strategy has not been well-characterized in standard practice.
Patients in a prone position tend to display improved oxygenation. The underlying mechanism includes a higher end-expiratory lung volume, improved ventilation of dependent regions (in part due to relief of the cardiac mass on lower lung lobes), improved ventilation-perfusion matching and, possibly decreased lung inhomogeneity. Prone positioning may reduce mortality among mechanically ventilated patients. Other complications associated with prone positioning, however, must be actively prevented, including airway dislodgement and obstruction, and decubitus ulcers.
Extracorporeal Membrane Oxygenation (ECMO)
ECMO may be instituted partially or totally. When partial ECMO is used, there is a reduced risk of complications from mechanical ventilation. The tidal volume requirement may also be less with the use of partial ECMO. The benefit and indications of ECMO, however, remain uncertain.
Neuromuscular blockade is useful in improving patient-ventilator synchronicity, especially among patients with ARDS who tend to display resistance against mechanical ventilation. This intervention may potentially reduce biotrauma, multi-organ dysfunction, and associated mortality.
The prophylactic use of anti-inflammatory agents and stem cells may be advantageous in reducing the incidence of biotrauma. However, this intervention is not well-characterized and is not routinely instituted in standard practice.
Oxygen delivery is determined by the fraction of inspired oxygen (FiO2) and the duration of oxygen delivery. Oxygen toxicity arises from the development of reactive oxygen species, namely superoxide, hydrogen peroxide, and the hydroxyl radical, which may cause cell damage, inflammation, and adverse genetic alterations. There has been no consensus yet regarding the safety threshold for inspired O2 (FiO2). The general recommendation is to use the lowest FiO2 that achieves adequate oxygenation.
Auto-positive End-expiratory Pressure (auto-PEEP)
Auto-PEEP is also known as "intrinsic PEEP" or "air-stacking." In healthy individuals, the opposing elastic force of the lungs equates that of the chest wall at the end of expiration at rest and, the end-expiratory lung volume (EELV) is approximately the same as the relaxation volume (V) of the airway. This accounts for the lack of end-expiratory airflow. Some mechanically ventilated patients, however, have an end-expiratory airflow above zero, despite the absence of persistent, active expiration.
Persistent end-expiratory airflow indicates that the EELV is greater than V. An increase in EELV above V constitutes pulmonary hyperinflation (a volume-based principle). Each tidal volume delivered during inspiration progressively increases the EELV and the end-inspiratory lung volume (EILV), resulting in air-stacking. These increases in lung volume cause progressive distension of the airway diameter. This, in turn, causes a reciprocal increase in end-expiratory elastic recoil of the lungs and chest wall (a pressure-based principle). The phenomenon is referred to as "auto-PEEP." The result is higher expiratory airflow.
Auto-PEEP is common among mechanically ventilated patients with comorbidities, such as asthma, chronic obstructive pulmonary disease (COPD); ARDS; acute respiratory failure; sepsis, and respiratory muscle weakness. Various causative, ventilator-mediated mechanisms have been identified, including (I) decreased expiratory time, (II) an increased expiratory time constant, (III) abnormally high minute ventilation, (IV) increased external airflow resistance, (V) an increased tidal volume, (VI) persistent inspiratory activity of respiratory muscles during expiration and (VII) increased lung compliance.
Respiratory Effects of Auto-PEEP
With auto-PEEP, there is a high EILV, which causes alveolar over-distension and decreased lung compliance. Since increased elastic recoil forces are required to overcome the distending forces, the elastic work of breathing effectively increases—the risk of barotrauma increases. There may be a requirement for more negative intrathoracic pressure that is equal to the end-expiratory, positive intra-alveolar pressure, to enable inspiratory airflow, i.e., there is also increased threshold work of breathing. Auto-PEEP may cause disturbances in respiratory muscle function. Additionally, hypoxia may ensue as a result of inhomogeneous distribution of inspired air between alveoli. Such disturbances may hinder efficient patient-ventilator interaction.
Cardiovascular Effects of Auto-PEEP
Auto-PEEP causes an increase in intrathoracic pressure. This results in decreased venous return and preload. Auto-PEEP also has the effect of increasing pulmonary vascular resistance. Thus, the right ventricular afterload is increased, impeding outflow. A more negative intrapleural pressure is required to enable air entry during inspiration; this may have the effect of increasing left ventricular afterload. The cumulative effect of these changes is a decrease in cardiac output. High positive pressure ventilation may also increase intrathoracic pressure to the extent that venous return from the head is reduced with a resultant rise in intracranial pressure. In this instance, delirium or agitation may ensue.
When auto-PEEP and associated hyperinflation are present, this appears on the ventilator monitor as persistent end-expiratory airflow. It may also be detected clinically from auscultation of a wheeze persisting to end-expiration. The presence of poor patient-ventilator interaction, pulsus paradoxus, fluctuations in pulse oximetry measurements, high plateau pressures, and development of hypotension are among other clinical cues to suspect the presence of auto-PEEP.
Therapeutic Strategies for Auto-PEEP
Therapeutic strategies should be targeted at the mechanism underlying auto-PEEP. Generally, interventions depend on the mode of mechanical ventilation (volume-controlled or pressure-controlled). Ventilator settings, such as respiratory rate, tidal volume, minute ventilation, inspiratory time and, heat moisture exchange filtration, may be adjusted to mitigate the effects of auto-PEEP. Ensuring adequate analgesia and temperature control are among supportive measures. Bronchodilator therapy may be instituted if clinically indicated. Strictly low-level PEEP application may be required. Intravascular fluid expansion may be instituted to maintain hemodynamic stability. Severe hypotension due to auto-PEEP may warrant prompt disconnection from the ventilator, followed by reconnection on clinical stabilization. Unresponsiveness of severe hypotension to ventilator disconnection should prompt suspicion of tension pneumothorax.
Overall, there is no single, standardized therapeutic intervention; treatment strategies need to be guided by patients' individual clinical requirements and risk profile.
Ventilator-associated Pneumonia (VAP)
Definition of VAP
In 2005, the American Thoracic Society (ATS) and Infectious Diseases Society of America (IDSA) jointly published guidelines on the management of patients with hospital-acquired pneumonia (HAP) and VAP, respectively. The guidelines were updated in 2016; however, the ATS and IDSA deemed the definitions of HAP and VAP presented in the 2005 guidelines clinically relevant and applicable. Therefore, they were not changed.
VAP is defined as the development of pneumonia more than 48 hours after the initiation of mechanical ventilation. To diagnose VAP, clinical features consistent with a diagnosis of pneumonia must not have been present prior to or at the time of intubation. Pneumonia has, in turn, been defined as the radiographic presence of new lung infiltrates and clinical evidence to suggest an infectious origin of thereof, including new-onset pyrexia, purulent sputum, leucocytosis or leucopenia and a decrease in oxygenation. Early-onset VAP is said to occur within the first 96 hours of mechanical ventilation and is associated with a better prognosis. Late-onset VAP manifests beyond the first 96 hours of mechanical ventilation and is associated with higher mortality and infection with multidrug-resistant (MDR) pathogens. VAP overall is associated with high morbidity and mortality.
HAP is a distinctly different entity to VAP. HAP is defined as pneumonia, not prevalent at the time of hospital admission, which develops two days or more after hospital admission. It may be prevalent even among those patients who are not mechanically ventilated.
Diagnosis of VAP
There is no gold standard for the diagnosis of VAP. The updated ATS and IDSA guideline (2016) on the management of HAP and VAP delineates "non-invasive sampling with semiquantitative cultures" and the use of clinical criteria as the recommended method to diagnose VAP.
Non-invasive respiratory sampling denotes endotracheal aspiration of respiratory secretions; a diagnostic culture result constitutes any amount of pathogenic microbial growth. (Invasive sampling entails blind bronchial sampling or employment of bronchoscopy for BAL or protected specimen brush (PSB) sampling techniques. In terms of quantification of culture results; the laboratory diagnostic cutoff values in (CFU/mL) is the same as outlined above in the surveillance section.
Additional clinical factors to consider in the diagnosis of VAP include the possibility of infection from an alternative source, the degree of clinical suspicion, the presence of sepsis, exposure to prior antimicrobial therapy at the time of specimen culture, and evidence of clinical improvement with antimicrobial therapy.
The Clinical Pulmonary Infection Score (CPIS) has been identified as a tool to elicit clinical features consistent with a diagnosis of VAP. Body temperature, leucocyte count and morphology, the presence of pathogenic bacteria on the culture of a respiratory specimen, the oxygen requirement (as gleaned from the ratio of the arterial partial pressure of oxygen to the fraction of inspired oxygen) and findings from chest radiography are some key components of the CPIS. A CPIS score of 6 or more (out of 12) indicates a high likelihood of VAP. Benefits of CPIS scoring include its usefulness for screening and its potential to enable early identification of VAP; however, it has limited sensitivity and specificity and is prone to interobserver variability.
Etiology of VAP
Of increasing concern is the incidence of VAP caused by MDR or extremely drug-resistant (XDR) pathogens, such as Pseudomonas aerugionsa, Acinetobacter species, Enterobacter species, Staphylococcus aureus and, Klebsiella pneumoniae. These are typically hospital-acquired and account for approximately 80% of all VAP cases.
Risk Factors for VAP
Risk factors for VAP include advancing age, comorbidity, prolonged length of stay in the ICU or hospital, prolonged duration of mechanical ventilation, and exposure to invasive procedures. The risk of VAP is also closely related to intubation and the prolonged presence of an endotracheal tube.
In the updated ATS and IDSA guideline (2016) on the management of HAP and VAP, the following are outlined as risk factors for MDR VAP: prior intravenous antibiotic treatment within 90 days, the presence of septic shock at the time of VAP diagnosis, ARDS preceding VAP, five or more days of hospitalization prior to the development of VAP and acute renal replacement therapy prior to the onset of VAP. Prior to antibiotic therapy and late-onset of symptoms are prominent risk factors for VAP due to methicillin-resistant Staphylococcus aureus (MRSA).
Treatment of VAP
Antimicrobial resistance, the risk of over-treatment, and the need to target specific, causative pathogens are major concerns in the context of treatment of VAP.
The guideline on the management of non-immunocompromised patients with HAP and VAP, as jointly published by the ATS and IDSA (2016), is evidence-based. However, it is intended only for voluntary use at the discretion of attending physicians and does not supersede physician assessment of individual patient requirements.
It is further recommended that local antibiotic programs be developed, which are specific to individual hospitals, ICUs, regions, or countries. These programs should be regularly updated and readily available to clinicians. The rationale is the significant variation in microbes and susceptibility patterns across different ICUs, hospitals, and geographic locations. Each antibiotic program should be guided by the commonly implicated microbial causes of VAP and by local susceptibility patterns to antimicrobial agents. Antibiotic programs should be evidence-based and governed by rational prescribing principles of safety, efficacy, suitability with respect to adverse effect profile, affordability, and availability, with due consideration to minimizing the development of antibiotic resistance.
Patients with clinically suspected VAP should receive empiric antimicrobial therapy that targets Staphylococcus aureus, Pseudomonas aeruginosa, and Gram-negative bacilli. Empiric antimicrobial coverage including two agents from different classes, targeted at Pseudomonas species, is reserved for patients with a high risk for antimicrobial resistance, for patients in an ICU setting where there is little information known regarding local antimicrobial susceptibility rates and, for patients where >10% of the Gram-negative microbial culture yield is resistant to monotherapy.
Inhaled antimicrobial therapy and other VAP treatment principles
Inhaled antimicrobial therapy, as an adjunct to systemic therapy, has been recommended for patients with VAP caused by Gram-negative bacilli susceptible only to aminoglycosides or polymyxins. Adjunctive, inhaled antimicrobial therapy may also be considered for patients not responding to systemic therapy alone, irrespective of the concern of generating microbial drug resistance. This recommendation is aimed at improving the rate of survival and places less priority on the cost implications of treatment.
The duration of antimicrobial therapy for patients with VAP is recommended as seven days. However, shorter or longer courses of therapy may be guided by the rate of improvement reflected by clinical, radiologic, and laboratory indices. Additionally, it is recommended that antimicrobial therapy be de-escalated, as opposed to fixed, i.e., VAP patients may be switched from a broad-spectrum antimicrobial regimen to a regimen targeted at the implicated pathogen, once culture results and susceptibility patterns have been determined. This may involve a change of antimicrobial agent or a change from combination therapy to monotherapy.
Clinical criteria and procalcitonin (PCT) levels may be used to guide discontinuation of antimicrobial therapy, although it is uncertain if PCT is actually a significantly beneficial indicator.
It is important to note that immunocompromised patients, by virtue of predisposition to opportunistic infections, would require a different management approach; some management principles would, however, be the same.
Healthcare Burden of VAP
In a retrospective analysis of mechanically ventilated patients enrolled in the NASCENT study, it was observed that median charges for patients with microbiologically confirmed VAP were nearly $200,000. Besides the hospital costs, it was also associated with a longer duration of intubation and hospital stay.
Gastrointestinal (GIT) Complications
Mechanisms of Colonization with Aerobic Gram-negative Bacteria (AGNB) and VAP
Mechanically ventilated patients are predisposed to gastric colonization with AGNB. This aspect contributes to the pathogenesis of VAP. In critical illness, there is impaired clearance of AGNB from the gastrointestinal (GI) tract. Additionally, reflux of intestinal contents into the stomach may occur, especially in the case of ileus. The presence of bilirubin in refluxed intestinal contents increases the gastric pH; a pH>4 favors gastric AGNB colonization. In patients on antimicrobial therapy, eradication of normal GI flora further impairs host defenses against AGNB colonization. Certain antimicrobial therapy, for example, with fluoroquinolones, has also been shown to promote the growth of fungal organisms in the stomach.
According to the "gastropulmonary hypothesis," potentially pathogenic microbes enter the stomach via an exogenous route (for example, through contaminated nasogastric feeding tubes) or via the endogenous reflux of intestinal content. Retrograde colonization of the oropharynx may ensue. Repeated micro-aspirations of oropharyngeal or gastric contents around the endotracheal tube cuff leads to colonization of the lower respiratory tract and consequent VAP. Comorbidities, such as diabetes mellitus and chronic liver disease, increase the risk for AGNB colonization of the GI tract. Colonization with AGNB may occur typically within seven days of admission to ICU and correlates with increasing disease severity and host immunocompromise.
Peptic Ulceration and Prophylaxis
Gastric colonization predisposes mechanically ventilated patients to peptic ulceration and associated upper gastrointestinal (GI) bleeding. Prevention strategies are, therefore, important, especially among patients with risk factors for severe bleeding events (for example, coagulopathy).
Peptic ulcer prophylaxis usually consists of histamine 2-receptor antagonists (H2-RA), which neutralize gastric acid or, antacids which reduce gastric acid secretion. Both H2-RA and antacids are effective in mitigating the risk of gastric ulceration and upper GI bleeding. However, they increase the gastric pH, effectively promoting AGNB colonization and increasing the risk of VAP. Hence, sucralfate is a suitable, alternative agent for peptic ulcer prophylaxis, as it does not increase the gastric pH. Additionally, it has cytoprotective and antimicrobial properties, which impede gastric bacterial colonization. Mechanically ventilated patients with risk factors for severe upper GI bleeding, including coagulopathy, should receive prophylaxis with an H2-RA rather than sucralfate.
Prevention of GIT Complications: Direct intestinal feeds
Enteral feeds generally have a pH between 6 and 7. The direct delivery of enteral feeds into the small intestine may lower the risk of aspiration and pneumonia by avoidance of the gastric pH-increasing effect of enteral feeds, which favors gastric AGNB colonization. However, the evidence surrounding this strategy has been conflicting, and it, therefore, does not feature in routine, standard practice.
Prevention of GIT complications: Selective Decontamination of the Digestive Tract (SDD)
SDD entails administration of 3 topical, non-absorbent antimicrobial prophylactic agents (polymyxin E, tobramycin or gentamycin, and amphotericin B) to the oropharynx to counteract the growth of AGNB. A nasogastric or orogastric tube is used for flushing the antimicrobial agents into the stomach. This is accompanied by a 3 to 4-day course of an intravenous third-generation cephalosporin at standard doses to cover for early, community-acquired infection with common organisms, such as Haemophilus influenza and Streptococcus pneumoniae.
SDD is usually initiated at admission to ICU and discontinued when mechanical ventilation is no longer required or, at discharge from ICU. The duration of SDD is determined by individual patient requirements; generally, it is administered for three days at 6-hourly intervals to achieve adequate decontamination.
SDD has emerged as a strategy to overcome impaired colonization resistance in critically ill patients. "Colonization resistance" refers to innate host defense mechanisms mediated by normal GI flora, which protect against gastric colonization with AGNB. These mechanisms include prevention of AGNB adherence to the GI epithelium, removal of AGNB from the GI tract, production of toxins that kill AGNB and, competitive consumption of nutrients to restrict the growth of AGNB.
SDD has proven beneficial in reducing the incidence of VAP in patients admitted to the ICU with medical, surgical, and trauma-related ailments. There has also been evidence to suggest that SDD reduces mortality, antibiotic requirement, length of stay in the ICU, and health care costs. SDD is especially useful in settings where the incidence of VAP is high, and antimicrobial resistance rates are low.
Notwithstanding the potential benefits, SDD is not implemented routinely in standard practice. A concern associated with its use is the potential for the selection of MDR organisms, notably Gram-positive bacteria. It also remains uncertain if the GI tract is the main source of pathogens typically implicated in the etiology of VAP.