VENTILATOR ASSOCIATED PNEUMONIA (VAP)

Transcription

1 VENTILATOR ASSOCIATED PNEUMONIA (VAP) Dr.Kolli S.chalam, MD; PDCC. Prof. & Head Dept. of Anesthesiology and Critical care Medicine, Sri Sathya Sai Institute of Higher Medical Sciences, white field, Bangalore Learning Objectives To provide an update on the incidence and potential consequences associated with ventilatorassociated pneumonia. To review current research on the topic of ventilator-associated pneumonia, including strategies for its treatment and prevention in critically ill patients. To suggest practices for the bedside clinician and also focuses on the controversies with regard to diagnostic tools and approaches, treatment plans, and prevention strategies. INTRODUCTION Ventilator-associated pneumonia is defined as parenchymal lung infection occurring more than 48 h after intubation and ventilation. VAP is the most common healthcare associated infection in intensive care unit. The condition is associated with increased morbidity, mortality, length of stay, and hence higher medical expenses. It accounts for about half of all antibiotics given in the intensive care unit (ICU). Lack of a gold standard definition leads to both under diagnosis and over diagnosis. VAP is a possibility if the ventilated patient has new or progressive radio-graphic infiltrate along with fever, purulent sputum, leukocytosis, and decline in oxygenation. A high clinical suspicion of pneumonia in a ventilated patient should prompt the immediate administration of an appropriate broad spectrum antibiotic(s). Implementation of evidence based interventions will reduce the incidence of pneumonia in all patients receiving mechanical ventilation. EPIDEMIOLOGY Incidence of VAP depends on case mix, duration of mechanical ventilation (early VS late onset VAP), and the diagnostic criteria used. It occurs in 9-27% of mechanically ventilated patients, with about five cases per 1000 ventilator days. VAP rate can be measured as follows: Total number of VAPs in all ICUs) / Total number of ventilator days in all ICUs x 1,000. Goal is to decrease the VAP rate per 1,000 ventilator days by 50 percent within 12 months.

2 The incidence of pneumonia has been reported to be 5-10 cases per 1000 hospital admissions, while in general ICU population incidence has ranged from 8 20%. The risk of nosocomial pneumonia increases by 6 to 20-folds in intubated patients. VAP affects between 20 and 70% of the ventilated patients and accounts for at least 10 episodes per 1,000 intubation days. The risk of pneumonia increases maximum during first five days. During first 5 days of ventilation, risk increases by 3%/day; during 5-10 days of ventilation it increases by 2%/day and after 10 days of ventilatory support it increases by 1%each day. In a recent PGI Chandigarh study, VAP was found to be 30.7% per 1,000 ventilator days. PATHOGENESIS Intubation and mechanical ventilation may impair mucociliary clearance and lead to sputum retention, airway occlusion, atelectasis, and VAP. Mucociliary clearance depends on the complex interaction between ciliated columnar epithelial cells of tracheobronchial tree and the special viscoelastic properties of the bronchial secretions. The mucociliary system represents an important protective mechanism of the upper and lower respiratory tract whereby inhaled particles and microorganisms are removed from the tracheobronchial system. Patients in the ICU have a tendency to retain secretions particularly while receiving long-term mechanical ventilation and, further, they are at the highest risk of developing pneumonia. The major potential reservoirs for organisms in a human body are stomach and sinuses. From these potential sites, the bacteria get entry into the trachea via aspiration of oropharyngeal pathogens or via leakage of bacteria around endotracheal tube cuff. From trachea, the microbial pathogens migrate to lower respiratory tract and here organisms are being colonized. Against the invasion entry or migration of micro bial pathogen, host defenses in the form of mechanical (ciliated epithelium and formation of mucus favoring expulsion of microorganism), in the form of humoral mechanism (formation of antibody and activation of compliment system) or in the form of cellular protection (through polymorphonuclear leucocyte, macrophages, lymphocytes and their cytokines) to expel the microor ganism out or to kill them or to make them inactive will play a crucial role. However if the defense mechanism is weak due to any factor or is insufficient to deal with the number of organisms, infection at lower respiratory tract takes place. APPROACH TO DIAGNOSIS The absence of a "gold standard" for VAP diagnosis is a major problem, with which diagnostic results can be compared. Diagnosing VAP requires a high clinical suspicion combined with bedside examination, radiographic examination, and microbiologic analysis of respiratory secretions. Although an etiologic diagnosis is made from a respiratory tract culture, colonization of the trachea precedes development of pneumonia in almost all cases of VAP, and thus a positive culture cannot always distinguish a pathogen from a colonizing organism. However, a sterile culture from the lower respiratory tract of an intubated patient, in the absence of a recent change in antibiotic therapy, is strong evidence that pneumonia is not present, and an extra pulmonary site of infection should be considered (Level II).

3 The diagnostic criteria of a radio graphic infiltrate and at least one clinical feature (fever, leukocytosis, or purulent tracheal secretions) have high sensitivity but low specificity (especially for VAP). Combinations of signs and symptoms may increase the specificity. When three clinical variables were used, the sensitivity declined, whereas the use of only one variable led to decline in specificity. All patients should have chest radiography, preferably postero-anterior. The radiography can help to define the severity of pneumonia and the presence of complications, such as effusions or cavitations. CT scan of chest further adds to confirmation. The clinical approach is overly sensitive, and it could be difficult to differentiate from other noninfectious conditions like congestive heart failure, atelectasis, pulmonary thromboembolism, pulmonary drug reactions, pulmonary hemorrhage, or ARDS. The clinical pulmonary infection score (CPIS) scoring system grades the severity of pneumonia and this includes six features. Each of these six features scores on a scale from 0 to 2, as follows: Temperature (Celsius) : >36.5 and <38.4 = 0, >38.5 and <38.9 =1, >39.0 or <36.5= 2 White Blood Cell Count : >4,000 and <11,000 = 0, <4,000 or >11,000 = 1, <4,000 or >11,000 and band forms >50% = 2 Tracheal Secretions : None or scant= 0; Non-purulent=1, Purulent =2 PaO2/FiO2 : 0= >240 or acute respiratory dis tress syndrome (ARDS), 2= 240 and no ARDS; Chest Radiograph : No infiltrate 0, Diffuse (or patchy) infiltrate = 1, Localized infiltrate =2 Pathogenic bacteria cultured from tracheal aspirate: rare or light quantity or no growth = 1 Moderate or heavy quantity (with same growth on Gram stain) = 1(1) CPIS score serves as tool to limit antibiotic abuse. VAP is less severe if CPIS score is less than or equal to 6 and prolonged course of antibiotics is unnecessary. Samples of lower respiratory tract secretions should be obtained from all patients with suspected VAP, and should be collected before antibiotic changes. Samples can include an endotracheal aspirate (EA), bronchoalveolar lavage (BAL) sample, or protected specimen brush (PSB) sample (Level II). A negative tracheal aspirate (absence of bacteria or inflammatory cells) in a patient without a recent (within 72 hours) change in antibiotics has a strong negative predictive value (94%) for VAP (Level II). Quantitative cultures of the homogenized EA and the presence of bacteraemia by blood culture were used. Several studies have suggested that the use of quantitative cultures of EA, a noninvasive & easily repeatable procedure, may have a similar diagnostic value compared with such invasive techniques as PSB and BAL which are expensive, time consuming, and require bronchoscopic procedure, a technique not always available throughout the day in the intensive care setting. Therefore, quantitative culture of EA can be an alternative to a more sophisticated testing in diagnosing VAP. One prospective study used a rapid immune-blot technique on BAL fluid in which infectious pneumonia suspected, and found that levels of soluble triggering receptor expressed on myeloid cells (strem-1) were the strongest independent predictor of pneumonia.

4 Being multifocal nature of VAP, BAL and endotracheal aspirates can provide more representative samples than the protected specimen brush (PSB), which samples only a single bronchial segment. Endotracheal aspirates (ETA) can be cultured quantitatively, and with a threshold of 10 6 cfu/ml or more, the sensitivity of this method for the presence of pneumonia has a mean of 76± 9% and specificity with a mean of 75±28%. Bronchoscopic BAL studies have typically used a diagnostic threshold of 10 4 cfu/ml. A review literature of 23 prospective studies of BAL in suspected VAP showed sensitivity with a mean of 73± 18%, and specificity with a mean of 82±19%. If bronchoscopic sampling is not immediately available, non bronchoscopic (NBAL)sampling can reliably obtain lower respiratory tract secretions for quantitative cultures, which can be used to guide antibiotic therapy decisions (Level II). MICROBIOLOGY The segregation of patients with VAP into groups categorizing them as early- and late-onset has been shown to be of paramount importance. Early-onset pneumonia, i.e., VAP 5 days commonly results from aspiration of endogenous community-acquired pathogens such as Enterobacteriaceae,Staphylococcus aureus, Streptococcus pneumoniae, Haemophilus influenza and Candida spp. with endotracheal intubation and impaired consciousness being the associated main risk factors. Late-onset pneumonia, i.e., VAP 5 days followed by aspiration of oropharyngeal or gastric secretions containing potentially drug-resistant nosocomial pathogens e.g. non-fermenters (Pseudomonas spp. and Acinetobacter spp.) were significantly associated. VAP is increasingly associated with MDR pathogens. Production of ESBL, AmpC beta-lactamases and metallo beta-lactamases were responsible for the multi-drug resistance of these pathogens. THERAPEUTIC CONSIDERATIONS Differentiations of different types of pneumonia are important as they differ in terms of predominance of causative organism, selection of antibiotic, and course and prognosis of the disease. To be more precise, the pneumonia has been broadly categorized as community acquired pneu monia (CAP) and hosp ital acquired pneumonia (HAP). Antibiotic therapy A major goal of therapy is eradication of the infecting organism, and so, antimicrobials are mainstay of treatment. Delays in the initiation of appropriate antibiotic therapy can increase the mortality of VAP and thus, therapy should not be postponed for the purpose of performing diagnostic studies in patients who are clinically unstable (level II).

5 The ATS has recently published guidelines to guide empirical antibiotic choices. These guidelines are divided into those for patients at risk for VAP caused by i) multidrug-resistant organisms and those for ii) patients without such risk. Risk factors for multidrug-resistant organisms include prior antimicrobial therapy in the preceding 90 days, current hospitalization exceeding 5 days (not necessarily ICU days), high frequency of resistance in the community or local hospital unit, and immunosuppressive disease and/or therapy. In the absence of risk factors for multidrug-resistant bacteria, the clinician should choose empirical therapy for Streptococcus pneumoniae, Haemophilus influenzae, methicillin-sensitive Staphylococcus aureus, and antibiotic-sensitive gram-negative enteric organisms. Antibiotic choices include ceftriaxone, quinolones (levofloxacin, moxifloxacin, or ciprofloxacin), ampicillin/sulbactam, or ertapenem Inappropriate therapy (failure of the etiologic pathogen to be sensitive to the administered antibiotic) is major risk factor for excess mortality and length of stay for patients with HAP, and antibiotic-resistant organisms are the pathogens most commonly associated with inappropriate therapy. When risk factors for multidrug-resistant organisms are present, the clinician must consider not only the organisms listed above but also Pseudomonas aeruginosa, Klebsiella, Enterobacter, Serratia, Acinetobacter, Stenotrophomonas maltophilia, Burkholderia cepacia, and methicillin-resistant S. aureus. Empirical therapy is broadened to include (i) either an antipseudomonal cephalosporin (cefepime or ceftazadime), an antipseudomonal carbepenem (imipenem or meropenem), or a β-lactam/β-lactamase inhibitor (pipercacillin-tazobactam) plus (ii) an antipseudomonal fluoroquinolone (ciprofloxacin or levofloxacin) or an aminoglycoside (amikacin, gentamicin, or tobramycin) plus linezolid or vancomycin. To achieve adequate therapy, it is necessary not only to use the correct antibiotic, but also the optimal dose and the correct route of administration (oral, intra venous, or aerosol) to ensure that the antibiotic penetrates to the site of infection, and to use combination therapy if necessary. Most b lactam antibiotics achieve less than 50% of their serum concentration in the lung, whereas fluoroquinolones and linezolid equal or exceed their serum concentration in bronchial secretions. There is a lot of debate over the use of antibiotics as monotherapy versus combination therapy. A meta analysis has evaluated all prospective randomized trials of b -lactam monotherapy compared with b lactam aminoglycoside combination regimens in patients with sepsis, of whom at least around 15% patients had either HAP or VAP, showed clinical failure was more common with combination therapy and there was no advantage in the therapy of P. aeruginosa infections, compared with monotherapy. In addition, combination therapy did not prevent the emergence of resistance during therapy, but did lead to a significantly higher rate of nephrotoxicity. Monotherapy with ciprofloxacin has been successful in patients with mild HAP (defined as a CPIS of 6 or less) but is less effective in severe HAP. Combination therapy should be used if patients are likely to be infected with MDR pathogens (Level II), though no data documented the superiority of this approach compared with monotherpay, except to enhance the likelihood of initially appropriate empiric therapy (Level I). If patients receive combination therapy with an aminoglycoside-containing regimen, the aminoglycoside can be stopped after 5-7 days in responding patients (Level III).

6 Monotherapy should be used and preferred over combination therapy whenever possible because combination therapy is often expensive and exposes patients to unnecessary antibiotics, thereby increasing the risk of MDR pathogens and adverse outcomes. Patients who develop nosocomial pneumonia with no risk factors for drug-resistant organisms are likely to respond to monotherapy with the antibiotics. Monotherapy is also the standard when gram-positive HAP, including MRSA, is documented. To use monotherapy in patients with severe VAP, the ATS committee believed that patients should initially receive combination therapy as described in, but therapy could be changed to a single agent if cultures did not shown a resistant pathogen. Patients in this risk group should initially receive combination therapy until the results of lower respiratory tract cultures are known and confirm that a single agent can be used (Level II). If P. aeruginosa pneumonia is documented, combination therapy is recommended, mainly because of the high frequency of development of resistance on monotherapy and also to avoid inappropriate and ineffective treatment of patients. In case of Acinetobacter species are to be present, the most active agents are the carbapenems, sulbactam, colistin, and polymyxin. There are no data documenting an improved outcome if these organisms are treated with a combination regimen. (Level II). If ESBL+ Enterbacteriaceae are present, then monotherapy with a third-generation cephalosporin should be avoided. The most active agents are the carbapenems. Duration of therapy If patients receive an initially appropriate antibiotic regimen, efforts should be made to shorten the duration of therapy from the traditional 14 to 21 days to periods as short as 7 days, provided that the etiologic pathogen is not P. aeruginosa, and that the patient has a good clinical response with resolution of clinical features of infection (Level I). Patients with a low clinical suspicion of VAP (CPIS of 6 or less) can have antibiotics safely discontinued after 3 days. Aerosolized antibiotics Aerosolized antibiotics have not been proven to have value in the therapy of VAP (Level I). But may be considered as adjunctive therapy with an inhaled amingolycoside or polymyxin for MDR gram negative pneumonia, especially in patients who are not improving with systemic therapy (Level III). More studies of this type of therapy are needed. Preference of particular antibiotic Linezolide is an alternative to vancomycin for the treatment of MRSA VAP (Level II) and may also be preferred if patients have renal insufficiency or are receiving other nephrotoxic agents, but more data are needed (Level III) Antibiotic rotation or restriction or holiday

7 Antibiotic restriction can limit epidemics of infection with specific resistant pathogens. Heterogeneity of antibiotic prescriptions, including formal antibiotic cycling, may be able to reduce the incidence of antibiotic resistance. But, the long-term impact of this practice is un known. (Level II). Continuous versus intermittent infusion Standard administration is by intermittent infusion; however, continuous infusion may be advantageous. Efficacy of drug with bactericidal activity increases with the exposure time that could be well maintained above minimum inhibitory concentration (MIC) by the continu ous infusion. However, sufficient evidence of its clinical efficacy is limited. SUPPORTIVE THERAPY Tracheostomy versus prolonged ETT Multicenter, randomized trial of mechanically ventilated patients with ongoing severe respiratory failure showed a trend towards lower VAP rates in patients who had tracheotomy at 1 week compared with tracheotomy at 2 weeks. However, the overall observed incidence of VAP was less than predicted. It did reduce the ventilator days. Chest physiotherapy, postural drainage, humidification and aerosolisation with bronchodilators, mucolytic agents are crucial but with little documentation of efficacy in the management of pneumonia. Multimodal physiotherapy: Effective suctioning is an essential aspect of airway management and has an important role to play in the prevention of VAP, especially early-onset VAP. Manual hyperventilation, percussion, vibration and postural drainage has shown better clearance of VAP. Non-pharmacologic measures Implementation of a microbiological surveillance protocol, good hand hygiene, minimization of the use of invasive devices. use of barrier methods to preventing cross-transmission a program of strict infection control that includes education of the healthcare team, appropriate Ratio of Health Personnel, the recommended nurse: patient ratio is 1:1 and also availability of skilled respiratory therapists Avoiding Unnecessary inpatient Transfers: the intra-hospital transfer of patients with MV increases the risk of developing VAP. Use of Non-invasive Ventilation (NIV) -Intubated patients have up to 21 times greater risk of acquiring nosocomial pneumonia than patients without an artificial airway. The guidelines of the

8 American Thoracic Society/Infectious Disease Society of America recommend NIV as an alternative for patients COPD etc. Early discontinuation from Mechanical Ventilation and proper weaning protocols to avoid extubation failures, preference of volume-controlled ventilation (VCV) over pressure-controlled ventilation (PCV) as VCV has decreased mortality rates. Oro gastric, oro-tracheal intubation versus nasotracheal to reduce the incidence of sinusitis and VAP. Prevention of Biofilm or Biolayer formation with silver-coated endotracheal tubes (SSD) drainage of Subglottic Secretions: meta-analysis have shown a decrease in the incidence of VAP, especially early onset, but not in mortality, Beware of the risk of tracheal injury in that area. Endotracheal Cuff Pressure Control: recommended to keep the endotracheal cuff balloon pressure between 20 and 30cmH 2 o to avoid leak. Prevent changes or manipulations of the Ventilator Tubing -: showed a higher incidence of VAP in patients whose ventilator circuits were changed every 24 hours as compared with those for whom the change was made at 48h. Advised change at 7d. HME can reduce the build-up of condensation and the colonisation of the tubing, but has not been associated with a decreased incidence of VAP. Inclined Position -inclined position (> 30 ), is recommended, especially in patients receiving enteral nutrition, Kinetic Beds: controlled clinical trials have shown a reduced risk of VAP in surgical and neurosurgical patients, but not in medical patients. Monitor gastric residual volumes to avoid over distension of stomach and then pulmonary aspiration. A recent meta-analysis of 11 controlled clinical trials conducted in the ICU, which evaluated mortality, the risk of aspiration and pneumonia, found no significant differences in the evolution and concluded that, in critically ill patients without evidence of gastrointestinal dysmotility, the use of post-pyloric nutrition not associated with clinical benefits. Non-pharmacologic measures have demonstrated an impact on the prevention of VAP. It is recommended to implement some of these interventions jointly to achieve better results. PREVENTION STRATEGIES

9 Prevention of VAP is given paramount importance in all quality control programs as it may help in improving clinical outcome and reduce costs. Moreover, there is an increased rate of VAP caused by multi-drug resistant strains in the recent years which may further add on to mortality and morbidity. To achieve this aim many measures have been evaluated and recommended. Incorporation of a set of evidence-based practices to prevent VAP, called the VAP bundles may reduce the incidence of VAP in mechanically ventilated patients. In addition, incorporation of these VAP bundles in the clinical practice may result in decreased ventilator days, ICU stay, and mortality rates. Components of VAP bundle: a % head elevation, b. Chlorhexidine mouth care c. Selective gut decontamination d. Stress ulcer prophylaxis e. Daily wake tests/sedation vacation f. Use of subglottic secretion drainage (SSD) endotracheal tube g. Closed suction systems (CSS) h. Heat and moisture exchangers (HMEs) i. Early weaning j. Hand washing Unfortunately, no large randomized, prospective or retrospective literatures on VENTILATOR ASSOCIATED PNEUMONIA, especially for adults are available in context of India. Guidelines provided by American Thoracic Society (ATS) VAP may develop from aspiration of oral secretions from area around the tube cuff. Hence, measures likely to reduce such aspiration like limiting the use of sedative and paralytic agents that depress cough and other host-protective mechanisms, maintaining adequate endotracheal cuff pressure (more than 20 cm H 2 O), and use of continuous aspiration of subglottic secretions, through specially designed endotracheal tubes, may all significantly reduce the incidence of VAP and are recommended by International guidelines. SSD is a simple measure that reduces chronic micro-aspirations from area around the cuff of endotracheal tubes and hence aid in VAP prevention. A large meta-analysis revealed that application of SSD reduces the incidence of VAP by nearly half, primarily by reducing early-onset pneumonia, reduces

10 the duration of mechanical ventilation by 2 days and the length of ICU stay by almost 3 days. In patients who developed VAP, use of SSD delayed the onset of VAP by 6.8 days. SOD (selective oral decontamination) with the use of oral antiseptic chlorhexidine was to be found with significantly reduction in rates of nosocomial infection. Now a days routine prophylactic use of antibiotics as a selective decontamination of the digestive tract (SDD) with cephalosporins, to reduce HAP, is avoided due to high level of antibiotic resistance. Use of CSS (closed suction system) to prevent VAP is a contentious issue. Although it may have some theoretical advantage over open suction system with less chance of external contamination, several meta-analytical reviews have failed to show any benefit in terms of reduced VAP rates, ICU length of stay or mortality. Passive humidifiers or heat-moisture exchangers reduce colonization of the ventilator circuit and hence may have a role in VAP prevention. ATS and Center for Disease Control (CDC) guidelines do not recommend use of HME filters for VAP prevention as their use has not consistently shown to reduce the incidence of VAP. Proper hand hygiene was perceived as an important measure to prevent VAP and was incorporated in their respective VAP bundles. H2 receptor blockers, sucralfate and proton pump inhibitors are most commonly employed agents for stress ulcer prophylaxis. Patients receiving ranitidine had a significantly lower rate of gastrointestinal bleeding than those treated with sucralfate. Presently, there is inadequate data regarding the role of proton pump inhibitors in VAP prevention. On the contrary, evidence suggests that these agents may increase the risk of Clostridium difficile disease, and hence, they are not currently recommended. Unnecessary liberal use of allogenic blood products and old stored blood are found to be a risk factor for increased incidence of HAP and in a prospective randomized trial comparing liberal and conservative "triggers" to transfusion in ICU patients not exhibiting active bleeding and without underlying cardiac disease demonstrated that awaiting a haemoglobin level of 7.0 g/dl as opposed to a level of 9.0 g/dl before initiating transfusion resulted in no adverse effects on outcome. Disease for example COPD, ARDS, burn, trauma, coma or impaired consciousness, multiple organ failure, longer sur gical procedure, thoracic or upper abdominal operations, immunosuppression (including systemic corticosteroids,) hypoalbuminemia, high APACHE II score and etc. Aggressive treatment of hyperglycemia has both theoretical and clinical support, but may not lead to significant benefit in patient with VAP. FUTURE DIRECTIONS New evidence on VAP preventive measures includes evidence for the efficacy of changes in endotracheal tube cuff design and materials, drainage of subglottic secretions, saline instillation prior to

11 tracheal suctioning, patient positioning, oral decontamination, aerosolized antibiotics, and probiotic use. In the absence of a clinical reference standard, the diagnosis of VAP remains problematic. Although extensive research on invasive sampling techniques for microbiological confirmation has been conducted, current evidence suggests that endotracheal aspirates are equivalent. Promising new diagnostic methods include non culture-based microbiological techniques and biomarkers. REFERENCES 1. John D Hunter.Ventilator associated pneumonia. BMJ 2012; American Thoracic Society (ATS) and Infectious Diseases Society of America (IDSA). Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcareassociated pneumonia. Am J Respir Crit Care Med 2005; 171: Sing et al. short course antibiotic therapy for patients with pulmonary infiltrate in ICU: Am J respire Crit Med 2000; 162: Pugin J. Clinical signs and scores (CPIS) for the diagnosis of ventilator-associated pneumonia. Minerva Anestesiol 2002; 68: Bouza E, Pérez MJ, Muñoz P, Rincón C, Barrio JM, Hortal J. Continuous aspiration of subglottic secretions in the prevention of ventilator-associated pneumonia in the postoperative period of major heart surgery. Chest 2008; 134: De SmetA M.G.A.et al.decontamination of the Digestive Tract and Oropharynx in ICU Patients. N Engl J Med 2009; 360: Siempos II, Vardakas KZ, Falagas ME. Closed tracheal suction systems for prevention of ventilator-associated pneumonia. Br J Anaesth 2008; 100: Fartoukh M, Maitre B, Honoré S, Cerf C, Zahar JR, Brun-Buisson C. Diagnosing pneumonia during mechanical ventilation: The clinical pulmonary infection score revisited. Am J Respir Crit Care Med 2003; 168: Terragni PP, Antonelli M, Fumagalli R, et al. Early vs late tracheotomy for prevention of pneumonia in mechanically ventilated adult ICU patients: a randomized controlled trial. JAMA. 2010; 303: