Noninvasive Ventilation in Postoperative Care of Lung Transplant Recipients

LUNG

Noninvasive Ventilation in Postoperative Care of Lung Transplant Recipients P. Feltracco, E. Serra, S. Barbieri, M. Milevoj, M. Furnari, S. Rizzi, F. Rea, G. Marulli, and C. Ori ABSTRACT Noninvasive positive pressure ventilation (NIPPV), which provides consolidated treatment of both acute and chronic respiratory failure, is increasingly being used in the postoperative care of lung transplant patients. Graft- and patient-related respiratory insufficiency requiring mechanical ventilation are common features in the postoperative period; they may persist for hours to days. Prolonged intubation, particularly in these immunocompromised patients, has been considered one of the main predisposing factors for developing nosocomial pneumonia. It has been associated with increased length of intensive care unit (ICU) stay as well. Noninvasive mechanical ventilation is nowadays an attractive choice to shorten weaning time and avoid reintubation following lung transplantation. Rapid extubation plus prompt NIPPV application is a useful strategy for lung recipients who do not completely fulfill the criteria for safe extubation. Unloading respiratory muscles, decreasing respiratory rate and sensation of dyspnea, improving ventilation/perfusion abnormalities, decreasing the heart rate, and improving hemodynamics are among the recognized benefits. Adding a noninvasive inspiratory support plus positive end-expiratory pressure (PEEP) to lung transplant recipients has been helpful to prevent airway injury and infections, avoiding the need for reintubation in cases of extubation failure, facilitating nocturnal sedation, treating the post-reimplantation syndrome and postoperative phrenic nerve dysfunction, and preventing reintubation in cases of readmission to the ICU. In our practice, the helmet system has emerged as the preferred interface; in cases of dyshomogeneous dorsobasal lung infiltrates, it allows effective ventilatory support in the prone position as well. OSTOPERATIVE RESPIRATORY FAILURE requiring mechanical ventilation is a common feature of single and bilateral lung transplantations. It may persist for hours to days. The decision to discontinue artificial invasive ventilation following lung transplantation is usually based on maintenance of satisfactory gas exchange along with adequate lung volumes and lack of respiratory muscle fatigue. However, in extremely debilitated recipients or

P

From the Departments of Pharmacology and Anesthesiology (P.F., E.S., S.B., M.M., M.F., S.R., C.O.) and Cardiothoracic Surgery (F.R., G.M.), University Hospital of Padua, Padua, Italy. Address reprint requests to Paolo Feltracco, Dipartimento di Farmacologia e Anestesiologia, Policlinico di Padova, Via Cesare Battisti 267, 35100 Padova, Italy. E-mail: paolofeltracco@ inwind.it

© 2009 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710

0041-1345/09/$–see front matter doi:10.1016/j.transproceed.2009.02.048

Transplantation Proceedings, 41, 1339 –1344 (2009)

1339

1340

patients who have received borderline grafts, prompt liberation from the ventilator may be challenging; the risk of unnecessary delay in extubation must be balanced against the risk of early reintubation. Prolonged intubation, particularly in immunocompromised patients, has been considered one of the main predisposing factors for developing nosocomial pneumonia.1 It has been associated with increased length of intensive care unit (ICU) and hospital stays. Patients requiring reintubation following an unsuccessful extubation have displayed greater mortality rates than those who were successfully extubated on the first attempt.2 Shortening weaning time and avoiding reintubation have become primary goals in the postoperative course of lung transplantation. Since 1995,3 noninvasive mechanical ventilation has been an attractive treatment in this setting. NONINVASIVE POSITIVE PRESSURE VENTILATION FOR PROMOTING EARLY EXTUBATION FOLLOWING LUNG TRANSPLANTATION

Hypoxemia and hypercapnia associated with tachypnea and muscle exhaustion are the main causes of unsuccessful weaning attempts following lung transplantation. Due to decreased overall activity and lack of coordination of respiratory muscles along with reduced diffusion capacity for oxygen in severely debilitated recipients, spontaneous T-piece breathing becomes almost ineffective. Increased dead space ventilation leads to frank hypoventilation and acidosis. Acidosis is in turn deleterious to muscle function and may contribute to the vicious circle of progressive pump failure. Postoperative endotracheal intubation is therefore prolonged to compensate for the increased systemic oxygen demand caused by spontaneous respiratory efforts and muscle fatigue. It has been demonstrated that infectious lung disease in intubated patients is mainly due to aspiration of pharyngeal secretions around the airway, rather than to contents breathed from the ventilator through the airway. Consequently, withdrawal from invasive ventilation becomes a crucial target in immunocompromised individuals. Early extubation protocols have been developed in lung transplantation. The strategies that have been proven effective to overcome the transition period to complete ventilatory autonomy include: epidural analgesia, early sitting position, physiotherapy, and noninvasive positive pressure ventilation (NIPPV) by facial mask or helmet. NIPPV in particular is considered a feasible mode of assisted ventilation to facilitate an early removal of mechanical ventilation. Benefits of NIPPV in assisting a difficult-to-wean patient or in treating impending muscle fatigue after apparently successful extubation have been demonstrated in several trials.4 Rapid extubation plus prompt NIPPV application is particularly suitable for lung recipients who do not completely fulfill the criteria for safe extubation. However, passing to noninvasive ventilatory assistance should only be considered after careful clinical judgment of graft characteristics. Appropriately trained staff must use optimally

FELTRACCO, SERRA, BARBIERI ET AL

individualized ventilator modes with close clinical and physiological monitoring for signs of treatment failure. Other than the early removal of the endotracheal tube, desirable goals of postoperative NIPPV include avoidance of hypercapnia and hypoventilation, a decreased respiratory rate and sensation of dyspnea, improved ventilation/perfusion abnormalities, decreased heart rate, and better hemodynamics. The acutely decreased intrathoracic pressure following extubation may cause increased venous return to the right ventricle and a possible leftward shift of the ventricular septum (ventricular interdependence). As a consequence, both pulmonary artery pressure and pulmonary artery occlusion pressure may increase due to contemporary right ventricular overfilling and reduced left ventricular diastolic compliance. Lung interstitial edema may occur because of pulmonary capillary overload and slow pulmonary vein drainage into the left atrium. If ventilatory distress after weaning originates from lung congestion and interstitial water, NIPPV plus appropriate levels of positive endexpiratory pressure (PEEP) are recommended to reduce extravascular lung water when diuretics alone are ineffective.5 The rationale for the use of noninvasive ventilatory support soon after “artificial airway” removal lies in its ability to offset the increased workload of the respiratory muscles and prevent alveolar derecruitment. Reduced muscular inspiratory capacity is typical of advanced end-stage lung disease. Pressure support delivered by a helmet or a mask appears to be as effective as invasive pressure support to prevent the loss of vital capacity and impede severe alveolar collapse. Elimination of the extra work caused by the endotracheal tube is a real advantage in these patients who exhibit limited tolerance to even slight increases in inspiratory resistance. By intermittently resting the inspiratory muscles, NIPPV enables hypercapnic and acidotic patients to decrease both the rate of respiratory drive and consumption of oxygen by the muscles. Noninvasive ventilatory methods are characterized by a more favorable intrathoracic hemodynamic balance than conventional invasive methods. In fact, lung inflation pressures are lower than those with classic mechanical ventilation.6 This is of great advantage in the postoperative period of the hemodynamically unstable patient such as the lung recipient. Less need for sedatives under NIPPV along with a more physiologic venous return may decrease the requirement for inotropes and vasoactive agents as well. Respiratory mechanics after lung transplantation may be negatively influenced by poor graft compliance associated with, and sometimes further aggravated by, the impairment of diaphragmatic function. Carrey et al7 have shown that NIPPV may also improve diaphragmatic electromyogram activity, decrease accessory muscle involvement, and reduce inspiratory muscle energy expenditure. Unlike in spontaneous difficult breathing, patients under NIPPV treatment show a slower and deeper breathing pattern which indicates improved alveolar ventilation. As a consequence, a stable

NONINVASIVE VENTILATION

rise in oxygenation and a reduction in resting hypercapnia have been frequently observed. It is well known that many lung transplant candidates are used to noninvasive ventilation (mainly nocturnal) while they are on the waiting list. This modality usually improves their daytime arterial blood gases and overall health, as well as reduces the severity and frequency of exacerbations of chronic lung disease.8 Previous experience with this modality makes postoperative NIPPV more acceptable and rather simple to apply. Postoperative assisted physiotherapy represents an essential aspect of a lung transplant program. However, muscle weakness in the presence of a wet stiff graft may impair the patient’s cooperation with the physiotherapist. Application of NIPPV during and after physiotherapy may help reduce the workload and oxygen consumption avoiding polypnea and CO2 retention. It also increases patient compliance and acceptance by facilitating fatigue relief. NIPPV IN THE PREVENTION OF AIRWAY INJURIES AND INFECTION

Upper and lower airway injuries and infections are among the most dreaded complications of lung transplantation, for they are associated with high morbidity and mortality rates. Mechanical ventilation after lung transplantation should last as short as possible, not only to decrease the chances of recognized parenchymal complications, but also to curtail negative effects on the new bronchial anastomosis. Dangerous inadvertent damage to the bronchial sutures may occur during blind aspiration or as a consequence of extremely elevated intrabronchial pressures in cases of insistent coughing or active expiration against the ventilator. The risk of barotrauma on airway sutures is particularly high within the first 2 weeks following lung transplantation. NIPPV is considered a nonaggressive method of ventilatory assistance; it is consequently increasingly proposed to avoid the negative effects of conventional invasive airway management. Early extubation followed by NIPPV shortens the period of potential blind airway management. An unnecessary delay in extubation has the potential to consistently increase the risk of respiratory tract infections. The endotracheal tube may damage the tracheal mucosa and thereby increase the susceptibility to microorganism invasion. Shortening the dependence on the endotracheal tube is particularly desirable for individuals with inflammation and impaired airway ciliary functions, common features following ischemia-reperfusion damage to the graft. By leaving the upper airway intact, NIPPV can reduce bacterial colonization and nosocomially acquired infections. Several studies9,10 have reported a lower rate of ventilatorassociated pneumonia (VAP) when mechanical ventilation is provided via NIPPV rather than through an endotracheal tube. In a study by Conti et al,11 NIPPV adopted for severe respiratory failure among chronic obstructive pulmonary disease (COPD) patients was compared with conventional mechanical ventilation. The NIPPV group experienced a lower rate of sepsis and septic shock. They showed a trend

1341

toward a lower incidence of nosocomial pneumonia during their time in the ICU. There are nowadays many randomized and observational reports on both immunocompetent and immunocompromised patients which demonstrate that noninvasive respiratory treatment drastically reduces the rates of infection and sepsis.11–13 In a prospective randomized study conducted on 40 solid organ transplant recipients, the use of NIPPV was associated with a trend toward fewer cases of VAP and a significant reduction in the incidence of severe sepsis, septic shock, length of ICU stay, and ICU mortality compared with standard treatment.14 Early application of NIPPV, which preserves airway defense mechanisms, may be particularly helpful to prevent pneumonia in high-risk patients such as immunocompromised lung recipients. NIPPV FOR EXTUBATION FAILURE

Respiratory failure after extubation may arise because of a “marginal” graft, pulmonary congestion or edema, atelectases, sputum retention, pneumonia, or acute organ rejection. The need for reintubation within hours or days after extubation is in itself a marker of increased severity of graft dysfunction, representing an independent risk factor for many postoperative complications.15 Muscle weakness accompanied by poor endurance in respiratory exercises and an ineffective cough prevent the recipient from maintaining adequate alveolar ventilation. Rapid shallow breathing ensues if graft compliance is markedly reduced. Prolonged “superficial” breathing leads to diffuse alveolar collapse in the dependent parts of the lung. Respiratory failure severe enough to require reintubation implies a greater risk for prolonged dependence on invasive airways. Fortunately, in clinical practice, many respiratory disorders following extubation may respond to specific interventions, such as fluid restriction, diuretics, bronchial toilette, cough induction, resolution of abdominal distension, and increased immunosuppression. They may be expected to improve over a period of hours or days. Increasing the efficiency of gas exchange and alleviating respiratory distress and fatigue with intermittent trials of noninvasive pressure support ventilation is always worth attempting in these circumstances. Postoperative hypoventilation along with residual depression of central respiratory drive and an increased tendency to retain bronchial secretions exacerbate basal graft collapse. This problem in turn reduces lung compliance and functional residual capacity and increases intrapulmonary shunting and FiO2 requirements. Unless the severity of parenchymal disease makes it clearly inappropriate, there is nothing to lose by trying NIPPV in cases of postextubation respiratory failure. If the recipient is awake, hemodynamically stable, and able to expectorate and protect the airway, even in the presence of moderate to severe respiratory distress with accessory muscle use and blood gas derangement, one should consider intermittent or prolonged NIPPV.

1342

Ventilatory insufficiency at any time following extubation should first be approached through a noninvasive method seeking to increase functional residual capacity, decrease atelectases and airway resistance, and improve ventilation/ perfusion matching. Almost every spontaneous breath can be supported if the patient is able to trigger the ventilator pressure support mode. By increasing the level of pressure support or PEEP, the time delay to activate the ventilator trigger can be significantly reduced, resulting in less work of breathing and oxygen demand. Application of NIPPV before respiratory distress becomes excessively severe may increase the end-expiratory lung volumes and prevent the progression of alveolar derecruitment. Even though the use of NIPPV to prevent the need for reintubation in patients who develop ventilatory failure has not led to definitive conclusions, recent observations have shown more favorable results. Nava et al16 have demonstrated that NIPPV was more effective than standard medical therapy (oxygen, aerosolized drugs, physiotherapy, etc) to prevent postextubation respiratory failure among a population at high risk for this complication. Patients treated noninvasively showed a lower rate of reintubation and a reduced risk of ICU mortality. At our institution, the implementation of a “prophylactic” NIPPV in all dyspneic lung recipients has led not only to a reduction in the reintubation rate, but also to a decreased need for a temporary tracheotomy. Avoiding a tracheotomy in the immediate postoperative period may help decrease the patient’s susceptibility to respiratory infections. Tracheotomy, in fact, interferes with swallowing, does not protect from gastroesophageal reflux, and may not prevent enteric bacteria from entering the upper airways.17 It should be noted that some patients hardly benefit from NIPPV even though it is properly applied. NIPPV inefficacy can result from the patient’s inability to handle copious purulent secretions, a deterioration in gas exchange despite full dependence on ventilatory support, a lack of clinical response because of severe disturbances of respiratory mechanics, alterations in mental status, or impending hemodynamic derangement. Under these circumstances, instituting NIPPV may be futile to avoid reintubation. Severely hypoxemic recipients with high oxygenation requirements, patients with diffuse, dense, patchy lung infiltrates, or those who have experienced bilateral infectious disease are suitable only for aggressive ventilatory support. NIPPV AS A TOOL TO FACILITATE NOCTURNAL SEDATION

Hypoventilation, owing to a reduction in both tidal volume and respiratory rate, is sometimes observed among these patients during sleep as well as after analgo-sedation administered to alleviate the discomfort of ICU procedures such as bronchoscopy, medications, or drainage positioning. Deterioration in ventilation may be considerable when nocturnal sedation is needed to promote a restoring sleep. Under these circumstances, the reduction in carbon dioxide

FELTRACCO, SERRA, BARBIERI ET AL

responsiveness and the increase in upper airway resistance may cause significant oxygen desaturation. Oxygen desaturation in turn promotes pulmonary vessel vasoconstriction and pulmonary hypertension, which can lead to right ventricle overload. In addition, nocturnal diaphragm elevation during light sedation in the near-supine position may be responsible for reduced ventilatory volumes and a greater tendency to collapse of the dorsobasal alveoli. Continuous positive airway pressure (CPAP) plus pressure support (PS) at night may prevent parenchymal derecruitment and favor reexpansion of microatelectatic areas. These factors could explain the maintenance of satisfactory arterial blood gases and ventilatory patterns observed among patients treated with NIPPV during sleep. Light to moderate sedation is not an absolute contraindication; NIPPV can be delivered successfully under close monitoring even in patients with a low level of consciousness.18 Nighttime rest is important in these individuals, since it contributes to an improved tolerance of daily activities and physiotherapy. NIPPV IN THE TREATMENT OF POST-REIMPLANTATION SYNDROME

Temporary graft dysfunction during the postoperative course may arise from the so-called post-reimplantation syndrome. This form of reperfusion injury is characterized by new radiographic findings and gas exchange abnormalities. It can persist for hours to days. The clinical picture includes pulmonary congestion, decreased lung compliance, diffuse or lobar patchy infiltrates, pulmonary hypertension, right ventricular failure, and respiratory fatigue. When the syndrome resolves only slowly, atelectatic areas frequently develop in the dependent parts of the grafts— dorsal and basal areas in the supine position. The aim of ventilatory management of this syndrome should be to maintain spontaneous breathing with active diaphragmatic movement. Noninvasive pressure support ventilation triggered by spontaneous inspiratory effort leads to a greater distribution of inspired gas to the dependent lung, thus increasing the aeration of dysatelectatic areas. Since pulmonary blood flow in the supine position remains distributed primarily to the dorsal regions, we managed to deliver NIPPV plus PEEP in the prone position; in so doing we somewhat circumvent this phenomenon. The observed improvement in oxygenation with NIPPV in the prone position may likely be attributed to the combined effects of a better ventilation/ perfusion match along the anteroposterior axis, redistribution of extravascular fluid, recruitment of nonaerated alveoli, and redirection of pulmonary blood flow.19 Trying to maintain assisted breathing and preventing further ventilatory-induced lung damage in nonrecruitable, poorly ventilated dorsal areas are factors of paramount importance. NIPPV IN CASES OF PHRENIC NERVE DYSFUNCTION

Temporary or permanent phrenic nerve damage is a rare but usually serious complication of the surgical procedure

NONINVASIVE VENTILATION

during lung transplantation. It is responsible for ventilatory complications in the postoperative period, often resulting in basal atelectases, prolonged mechanical ventilation, difficult weaning, delayed extubation, and tracheostomy. Diaphragmatic plication generally achieves incomplete results. NIPPV can be a valid ventilatory option for cases of phrenic nerve dysfunction both to accelerate weaning from the ventilator and to continue ventilatory assistance until definitive autonomy. Berk et al.20 have demonstrated that in cases of postoperative phrenic nerve dysfunction, the use of noninvasive respiratory support shortens the duration of mechanical ventilation and decreases the length of ICU stay. NIPPV FOR VENTILATORY ASSISTANCE IN CASES OF READMISSION TO THE ICU

Respiratory failure in immunocompromised lung transplant recipients is one of the main causes of readmission to the ICU. The need for intensive treatment may occur at any time during their posttransplantation course. Muscle atrophy due to poor nutritional status and steroid treatment causes lung recipients to experience reduced respiratory reserve. Acute lung failure may be the result of infectious disease, acute cardiogenic pulmonary edema with interstitial and alveolar flooding, airway obstruction, pleural effusion, organ rejection, etc. Because of severely impaired gas exchange, mechanical ventilation represents an indispensable therapy for restrictive and/or obstructive patterns and illness-inducing muscular weakness. However, compared with other groups of medical patients receiving invasive ventilation, slightly worse survival is observed among lung transplant patients requiring mechanical ventilation in the ICU.21 Considering that it has become a treatment of choice in and outside the ICU for patients with acute exacerbation of chronic respiratory disease, the use of NIPPV should be recommended as well for respiratory insufficiency of lung transplant patients. It should be perceived as a way to possibly avoid endotracheal intubation rather than as an absolute alternative to invasive airway management. In fact, at the moment there are no absolute criteria to predict its success in patients admitted to the ICU with moderate to severe respiratory distress. Generally, patients with atelectases, cardiogenic pulmonary edema, acute lung injury (ALI), and early-stage posttransplant pneumonia seem to respond better than recipients with frank adult respiratory syndrome (ARDS) or diffuse pneumonia who often worsen after an initial phase of improvement. In a failing patient it must be viewed as a simple method of ventilatory assistance at a stage when an endotracheal tube is not yet necessary. In these patients, the combined positive effects of PEEP and inspiratory support may reopen atelectatic alveoli, increase lung ventilation, facilitate diaphragm excursion, and improve lung diffusion capacity. Rocco et al22 reported that the administration of NIPPV in patients developing acute respiratory failure following lung transplantation was successful to improve

1343

gas exchange, thereby preventing intubation in the large majority of subjects. In our experience, NIPPV was well tolerated without major complications even when applied in late phases of respiratory failure. EQUIPMENT SELECTION AND INTERFACE

Noninvasive ventilatory support can be provided by standard ICU ventilators, bilevel positive airway pressure (BiPAP), machines, portable volume, and pressure ventilators. Volumecontrolled, pressure-controlled, and pressure support modes have proven beneficial. With these devices airway pressure delivered during inspiration is usually associated with end-expiratory pressure. By increasing mean airway pressure, respiratory unloading may be maximized.23 Timecycled or flow-cycled ventilators are generally both suitable for noninvasive support. The initial inspired pressure should be set to the highest level tolerated and then increased gradually over hours. Peripheral pulse oximetry must be included as a minimum standard of monitoring. Various interfaces have become available for delivering NIPPV. Oral, oronasal, and nasal masks along with the helmet have been successfully applied. Description of the advantages and disadvantages of each device has been reported elsewhere.24 Both the facial mask and the helmet allow great flexibility in applying and removing ventilatory assistance. In our practice, in accord with patient comfort and personnel preferences, we prefer the helmet system. In the presence of low graft compliance, the helmet is suitable for delivering high inflation pressures even for extended periods of time. Through clinical education and practice, respiratory therapists and nurses have noticed that patients are more compliant to execute respiratory exercises using the helmet rather than the facial mask. Major reported concerns regarding its use are the risk of rebreathing and potential problems with effective triggering and cycling of the ventilator, because of the large gas volume within the helmet. Improvements in manufacturing, proper selection of the device, and well-individualized ventilator settings have nowadays reduced air leaks, provided better synchronization, and avoided rebreathing. The helmet system allows prolonged continuous application of NIPPV. As reported by Rocco et al,25 immunocompromised patients treated with the helmet because of acute respiratory failure required a significantly lower average number of discontinuations of noninvasive treatment compared with those treated by mask. More patients in the helmet group showed sustained improvements in PaO2/FiO2 ratios over time even after helmet discontinuation. Conti et al26 showed that use of the helmet in patients with postoperative respiratory failure was better tolerated and showed a significantly lower rate of NIPPV failures than the face mask. Noninvasive CPAP by helmet may provide further benefits once the recipient has been discharged to the ward. CPAP is, in fact, easily delivered outside the ICU by simple continuous flow systems without the need for a mechanical ventilator. Proper devices to noninvasively deliver CPAP along with

1344

improved expertise of “outside the ICU staff” make CPAP well accepted and safe for transient episodes of oxygen desaturation. In conclusion, NIPPV is increasingly utilized to support acute and chronic respiratory disorders in various settings. Evidence of efficacy in assisting ventilatory failure of immunocompromised patients is rapidly accumulating. Shortening the duration of invasive mechanical ventilation and preventing reintubation are great benefits in the postoperative care of lung transplant recipients, even though, at present, evidence-supported indications are scarce. Whether these modalities have an effective role in reducing short-term postoperative complications, overall hospital stay, and mortality in these subjects remains to be confirmed. When adopted to support a failing recipient soon after extubation, or for early treatment of impending respiratory failure at any time posttransplantation, NIPPV has been associated with a high success rate and minimal treatment-related complications. Considered a “gentle” method of ventilation which avoids “invading” the upper airway, it is recommended not only to reduce postoperative VAP, but also to prevent traumatic complications in the new graft as well. Incorporation of NIPPV into postoperative care of lung transplant patients requires a concerted effort by physicians, respiratory therapists, and nurses. Coaching the recipients, individualizing specific needs, assuring patient compliance, and monitoring efficiency are among the recommendations for successful implementation. REFERENCES 1. Chastre J, Fagon JY: Ventilator-associated pneumonia. Am J Respir Crit Care Med 165:867, 2002 2. Epstein SK, Ciubotaru RL, Wong JB: Effect of failed extubation on the outcome of mechanical ventilation. Chest 112:186, 1997 3. Kilger E, Briegel J, Haller M, et al: Noninvasive ventilation after lung transplantation. Med Klin 90:26, 1995 4. Ferrer M, Esquinas A, Arancibia F, et al: Noninvasive ventilation during persistent weaning failure: a randomized controlled trial. Am J Respir Crit Care Med 168:70, 2003 5. Gust R, Gottschalk A, Schmidt H, et al: Effects of continuous (CPAP) and bi-level positive airway pressure (BiPAP) on extravascular lung water after extubation of the trachea in patients following coronary artery bypass grafting. Intensive Care Med 22:1345, 1996 6. Celikel T, Sungur M, Ceyhan B, et al: Comparison of non-invasive pressure ventilation with standard medical therapy in hypercapnic acute respiratory failure. Chest 114:1636, 1998 7. Carrey Z, Gottfried SB, Levy RD: Ventilatory muscle support in respiratory failure with positive pressure ventilation. Chest 97:150, 1990 8. Wedzicha JA, Muir JF: Noninvasive ventilation in chronic obstructive pulmonary disease, bronchiectasis and cystic fibrosis. Eur Respir J 20:777, 2002

FELTRACCO, SERRA, BARBIERI ET AL 9. Carlucci A, Richard JC, Wysocki M, et al: SRLF Collaborative Group on Mechanical Ventilation. Noninvasive versus conventional mechanical ventilation: an epidemiologic survey. Am J Respir Crit Care Med 163:874, 2001 10. Girou E, Schortgen F, Delclaux, et al: Association of noninvasive ventilation with nosocomial infections and survival in critically ill patients. JAMA 284:2361, 2000 11. Conti G, Antonelli M, Navalesi R, et al: Noninvasive vs conventional mechanical ventilation in patients with chronic obstructive pulmonary disease after failure of medical treatment in the ward: a randomized trial. Intensive Care Med 28:1701, 2002 12. Nourdine K, Combes P, Carton MJ, et al: Does non-invasive ventilation reduce the ICU nosocomial infection risk? Intensive Care Med 25:567, 1999 13. Hilbert G, Gruson D, Vargas F, et al: Noninvasive ventilation in immunosuppressed patients with pulmonary infiltrates, fever and acute respiratory failure. N Engl J Med 344:481, 2001 14. Antonelli M, Conti G, Bufi M, et al: Noninvasive ventilation for treatment of acute respiratory failure in patients undergoing solid organ transplantation. JAMA 283:235, 2000 15. Torres A, Gatell JM, Aznar E, et al: Re-intubation increases the risk of nosocomial pneumonia in patients needing mechanical ventilation. Am J Respir Crit Care Med 152:137, 1995 16. Nava S, Gregoretti C, Fanfulla F, et al: Noninvasive ventilation to prevent respiratory failure after extubation in high-risk patients. Crit Care Med 33:2465, 2005 17. Elpen EH, Scott MG, Petro L, et al: Pulmonary aspiration in mechanically ventilated patients with tracheostomies. Chest 105: 563, 1994 18. Diaz GO, Alcaraz AC, Talavera JC, et al: Noninvasive positive-pressure ventilation to treat hypercapnic coma secondary to respiratory failure. Chest 127:952, 2005 19. Guerin C, Badet M, Rosselli S, et al: Effects of prone position on alveolar recruitment and oxygenation in acute lung injury. Intensive Care Med 25:1222, 1999 20. Berk Y, van der Bij W, Erasmus ME, et al: Non-invasive ventilation in phrenic nerve dysfunction after lung transplantation: an attractive option. J Heart Lung Transplant 25:1483, 2006 21. Hadjiliadis D, Steele MP, Govert JA, et al: Outcome of lung transplant patients admitted to the medical ICU. Chest 125:1040, 2004 22. Rocco M, Conti G, Antonelli M, et al: Non-invasive pressure support ventilation in patients with acute respiratory failure after bilateral lung transplantation. Intensive Care Med 27:1622, 2001 23. Appendini L, Patessio A, Zanaboni S, et al: Physiologic effects of end-expiratory pressure and mask pressure support during exacerbations of chronic obstructive lung disease. Am J Respir Crit Care Med 149:1069, 1994 24. Hess RD: The evidence for non-invasive positive-pressure ventilation in the care of patients in acute respiratory failure: a systematic review of the literature. Respir Care 49:810, 2004 25. Rocco M, Dell’Utri D, Morelli A, et al: Noninvasive ventilation by helmet or face mask in immunocompromised patients. Chest 126:1508, 2004 26. Conti G, Cavaliere F, Costa R, et al: Noninvasive positivepressure ventilation with different interfaces in patients with respiratory failure after abdominal surgery: a matched-control study. Respir Care 52:1463, 2007