Noninvasive Ventilation for Critical Care* Erik Garpestad, MD, FCCP; John Brennan, MD; and Nicholas S. Hill, MD, FCCP

CHEST Noninvasive Ventilation for Critical Care* Erik Garpestad, MD, FCCP; John Brennan, MD; and Nicholas S. Hill, MD, FCCP Postgraduate Education Corner CONTEMPORARY REVIEWS IN CRITICAL CARE MEDICINE Noninvasive ventilation (NIV), the provision of ventilatory assistance without an artificial airway, has emerged as an important ventilatory modality in critical care. This has been fueled by evidence demonstrating improved outcomes in patients with respiratory failure due to COPD exacerbations, acute cardiogenic pulmonary edema, or immunocompromised states, and when NIV is used to facilitate extubation in COPD patients with failed spontaneous breathing trials. Numerous other applications are supported by weaker evidence. A trial of NIV is justified in patients with acute respiratory failure due to asthma exacerbations and postoperative states, extubation failure, hypoxemic respiratory failure, or a do-not-intubate status. Patients must be carefully selected according to available guidelines and clinical judgment, taking into account risk factors for NIV failure. Patients begun on NIV should be monitored closely in an ICU or other suitable setting until adequately stabilized, paying attention not only to vital signs and gas exchange, but also to comfort and tolerance. Patients not having a favorable initial response to NIV should be considered for intubation without delay. NIV is currently used in only a select minority of patients with acute respiratory failure, but with technical advances and new evidence on its proper application, this role is likely to further expand. (CHEST 2007; 132:711 720) Key words: acute respiratory failure; COPD; mechanical ventilation; noninvasive ventilation Abbreviations: ALI acute lung injury; APACHE acute physiology and chronic health evaluation; CHF congestive heart failure; CI confidence interval; CPAP continuous positive airway pressure; CPE cardiogenic pulmonary edema; DNI do not intubate; Fio 2 fraction of inspired oxygen; NIV noninvasive ventilation; PEEP positive end-expiratory pressure One of the most important developments in the field of mechanical ventilation over the past 15 years has been the emergence of noninvasive ventilation (NIV) as an increasing part of the critical care armamentarium. Although similar techniques such *From the Division of Pulmonary, Critical Care, and Sleep Medicine, Tufts-New England Medical Center, Boston, MA. Dr. Hill has received honoraria and research grants from and served on the medical advisory boards of ResMed, Inc., and Respironics, Inc. He has received a research grant from Versamed, Inc. Drs. Garpestad and Brennan have no conflicts of interest to disclose. Manuscript received October 28, 2006; revision accepted March 12, 2007. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Nicholas S. Hill, MD, FCCP, Division of Pulmonary, Critical Care and Sleep Medicine, Tufts-New England Medical Center, 750 Washington St, Boston, MA 02111; e-mail: nhill@tufts-nemc.org DOI: 10.1378/chest.06-2643 as intermittent positive pressure breathing were used widely during previous decades, unlike NIV they were used mainly to provide intermittent aerosol therapy. The term NIV includes other forms of ventilatory assistance that avoid airway invasion, such as negative pressure ventilation, but the vast majority of NIV applications now use positive pressure. Noninvasive application of continuous positive airway pressure (CPAP) will be considered a form of NIV here when used to treat certain types of respiratory failure, but it is not a true form of ventilatory assistance because the positive pressure does not increase intermittently to assist inspiration. The emergence of NIV has been fueled by its relative ease of application compared to alternative forms of noninvasive ventilation, as well as its demonstrated ability to improve patient outcomes in certain forms of acute respiratory failure compared to previously standard therapy, including endotra- www.chestjournal.org CHEST / 132 / 2/ AUGUST, 2007 711

cheal intubation. 1 This update will focus on recent developments regarding acute applications of NIV, including the expanding evidence base, technical advances, and assessment of current utilization. We emphasize techniques for proper patient selection and implementation that are critical if success rates reported in the literature are to be duplicated. NIV for Acute Respiratory Failure Recommended Indications Many applications of NIV have been tried in the critical care setting, but as of yet, only four are supported by multiple randomized controlled trials and metaanalyses. COPD Exacerbations The strongest level of evidence, including multiple randomized controlled trials, 2 7 supports the use of NIV to treat exacerbations of COPD. Also, metaanalyses by Ram et al 8 and Keenan et al 9 demonstrate more rapid improvements in vital signs and gas exchange as well as reductions in the need for intubation (relative risk, 0.41; 95% confidence interval [CI], 0.33 to 0.53; risk reduction, 28%), decreased mortality (relative risk, 0.52; 95% CI, 0.35 to 0.76; risk reduction, 10%), and decreased hospital length of stay ( 3.24 days; 95% CI, 4.42 to 2.06 days and 4.57 days, respectively). The Cochrane analysis 8 also noted more rapid improvements in vital signs, ph, and gas exchange, and reduced complication rates and hospital lengths of stay. Based on these observations, NIV should now be considered the ventilatory modality of first choice to treat acute respiratory failure caused by exacerbations of COPD. Acute Cardiogenic Pulmonary Edema Similarly strong evidence supports the use of noninvasive positive pressure techniques to treat acute cardiogenic pulmonary edema (CPE). 10 17 Recent metaanalyses 18 20 on the use of NIV to treat acute pulmonary edema have shown that both CPAP and NIV lower intubation and mortality rates compared to conventional therapy with oxygen, although the reduction in mortality rate was statistically significant only in one of the metaanalyses. 20. A randomized trial 17 comparing CPAP directly to NIV showed no difference in outcomes between the two to treat CPE, a finding confirmed in a recent metaanalysis by Ho and Wong. 21 Accordingly, by virtue of its greater simplicity and lesser expense, CPAP has been suggested as the initial noninvasive choice for acute CPE. However, some studies 22 have observed more rapid improvements in gas exchange and vital signs with NIV than with CPAP alone, so NIV may be preferable for patients with persisting dyspnea or hypercapnia after initiation of CPAP. Facilitating Extubation in COPD Patients Another NIV application supported by multiple randomized trials is to facilitate extubation in COPD patients. Candidates for early extubation are those who were intubated for COPD exacerbations because they were poor candidates for or failed NIV initially and are unable to pass a T-piece trial even though they have improved sufficiently to tolerate NIV. Ferrer et al 23 confirmed earlier findings of Nava et al 24 in such patients, randomizing 43 patients with persistent weaning failure (failure of three consecutive T-piece trials) to be extubated to NIV or weaned using conventional methods. They observed that NIV-treated patients had shorter durations of intubation (9.5 days vs 20.1 days) and ICU (14 days vs 25 days) and hospital stays (14.6 days vs 40.8 days), decreased incidence of nosocomial pneumonia (24% vs 59%), and improved ICU and 90-day survivals (80% vs 50%) [all p 0.05]. These studies strongly support the use of NIV to facilitate extubation in patients with hypercapnic respiratory failure and to avoid the complications of prolonged intubation. But it must be applied cautiously: only in patients who are otherwise good candidates for NIV and were not difficult intubations. Immunocompromised Patients The use of NIV is also well supported for immunocompromised patients who are at high risk for infectious complications from endotracheal intubation, such as those with hematologic malignancies, AIDS, or following solid-organ or bone marrow transplants. In a randomized trial 25 of patients with hypoxemic respiratory failure following solid-organ transplantation, NIV use decreased intubation rate (20% vs 70%, p 0.002) and ICU mortality (20% vs 50%, p 0.05) compared with conventional therapy with oxygen. Hilbert et al 26 observed fewer intubations (46% vs 77%) and a lower mortality rate (50% vs 81%) [both p 0.05] among immunocompromised patients (mainly hematologic malignancies, but some after solid-organ transplantation or with AIDS) with acute respiratory failure randomized to NIV as opposed to conventional therapy. The sizable reductions in mortality in these studies strongly support the early use of NIV as the initial ventilatory modality in immunocompromised patients with acute respiratory failure, although morbidity and mortality rates are still likely to be high. 712 Postgraduate Education Corner

Conditions for Which NIV Should Be Considered These applications are supported by a single randomized controlled trial, historically controlled trials, or multiple trials yielding conflicting evidence. NIV can be tried if patients are selected and monitored carefully. Asthma Several uncontrolled series 27 29 and one randomized trial 30 support the use of NIV for acute asthma. In the randomized trial, 30 a pilot, 33 patients with acute dyspnea but not in respiratory failure were randomized to standard therapy with nasal bilevel ventilation for3horstandard therapy with sham NIV. NIV improved expiratory flow rates more rapidly (80% of patients had a 50% increase in FEV 1 in the first hour, compared to only 20% in the sham group) and reduced the need for hospitalization. 30 The authors speculated that the positive pressure had a salutary effect on airway dilatation, although these results have yet to be replicated. A case report 31 raised concerns about the use of NIV for asthma, and a recent metaanalysis by Ram et al 32 concluded that routine use of NIV in severe acute asthma could not be recommended. We believe that a cautious trial should still be considered in asthmatics not responding to the first hour of conventional therapy, but more study is warranted. Postoperative Respiratory Failure Either NIV or CPAP may be helpful in averting postoperative respiratory failure by preventing atelectasis and/or improving gas exchange as suggested by three randomized controlled trials in patients undergoing different surgical procedures. Following thoracoabdominal aneurysm repair, prophylactic use of CPAP reduces overall pulmonary complications. 33 Squadrone et al 34 compared CPAP vs conventional oxygen therapy in patients with hypoxemic respiratory failure after major elective abdominal surgery. The CPAP group had lower rates of intubation, pneumonia, and sepsis. In the only randomized, controlled trial of NIV in postoperative patients, Auriant et al 35 found that NIV reduced intubation and mortality rates in patients with hypoxemic respiratory failure following lung resection. These trials indicate that either CPAP or NIV should be strongly considered to prevent or treat postoperative respiratory failure, mainly after lung resection in patients with underlying COPD or congestive heart failure (CHF). Although multiple studies support this application, further studies need to focus on the use of NIV following specific surgical procedures before firmer recommendations can be made. Do-Not-Intubate Patients Use of NIV for patients with acute respiratory failure who have a do-not-intubate (DNI) status has aroused debate. Concerns have been raised that the modality might merely prolong the dying process while mask discomfort outweighs any palliative effect. 36 However, a prospective observational study by Levy et al 37 showed that patients with reversible diagnoses such as COPD and CHF had a betterthan-even chance of surviving the hospitalization if treated with NIV (52% and 75%, respectively), whereas those with pneumonia or cancer had much lower likelihoods of hospital survival. Schettino et al 38 reported similar findings in their observational cohort and noted low success rates for NIV in postextubation respiratory failure, hypoxemic respiratory failure, and end-stage cancer. Some have proposed resolving the conflict about the use of NIV in DNI patients by specifying the goals of therapy. 39 Patients with reversible processes such as COPD exacerbations or CHF may wish to survive the acute illness and thus use NIV as a form of life support. They are willing to endure some discomfort to achieve this aim. Others desire palliation, aiming to alleviate dyspnea or briefly prolong survival to settle affairs. In these latter circumstances, excessive mask discomfort would justify stopping therapy because the goal of palliation is not being met. Differentiating between and agreeing on these aims requires close and effective communication between caregivers, patient and family. Hypoxemic Respiratory Failure Randomized controlled trials suggest that patients with hypoxemic respiratory failure (ie, severe respiratory distress, Pao 2 /fraction of inspired oxygen (Fio 2 ) 200 and a non-copd cause for respiratory failure) benefit from use of NIV. 40,41 In 105 such patients, Ferrer et al 41 found that compared to conventional therapy, NIV reduced the intubation rate (52 to 25%), the incidence of septic shock (31 to 12%) and ICU mortality (39 to 18%), and improved 90-day mortality (all p 0.05). Notably, almost one third of the patients had CPE, which would be expected to respond favorably to NIV, but patients with pneumonia were the ones that benefited the most in this study. This latter finding contrasts with earlier trials that showed an association between pneumonia and NIV failure 42 44 and the need for intubation in approximately two thirds of patients with pneumonia treated initially with NIV. 45 One randomized controlled trial 46 studying patients with severe communityacquired pneumonia showed that NIV improved outcomes including survival at 2 months but only in www.chestjournal.org CHEST / 132 / 2/ AUGUST, 2007 713

patients with underlying COPD. In one prospective cohort study, 47 risk factors for NIV failure in patients with acute hypoxemic respiratory failure included the diagnoses of ARDS or severe community-acquired pneumonia, more severe hypoxemia (Pao 2 / Fio 2 146 after the first hour of treatment), and age 40 years. In another study 48 of NIV to treat hypoxemic respiratory failure, shock, severe hypoxemia, and severe metabolic acidosis were associated with poor outcomes. In a prospective analysis 49 of selected patients with acute lung injury (ALI)/ARDS treated with NIV as the initial ventilator modality, a simplified acute physiology score II 34 and a Pao 2 /Fio 2 175 after 1 h of therapy predicted failure. The outcome of NIV is also very poor when used to treat hypoxemic respiratory failure in patients with idiopathic pulmonary fibrosis, 48 and this application should be discouraged. A concern that must be stressed is that the diagnostic category of hypoxemic respiratory failure is very broad and benefits accruing to certain subsets of patients within the larger diagnostic category could obscure adverse consequences in smaller subgroups. Thus, with the exception of use for CPE, which is supported by strong evidence, NIV should be used only with caution in carefully selected patients with hypoxemic respiratory failure, and those at high risk for failure should be considered for early intubation, especially if oxygenation fails to improve substantially within the first hour or two (Fig 1). Extubation Failure The use of NIV to prevent or treat extubation failure has also raised concerns. Respiratory failure following extubation imparts a poor prognosis; the duration of mechanical ventilation is lengthened, the likelihood of discharge to a chronic care facility is increased, and mortality may reach 40%. 50 Esteban et al 51 evaluated the ability of NIV to avoid extubation failure by randomizing patients to NIV or conventional therapy if risk factors developed, including hypercapnea, hypoxemia, acidemia, or tachypnea, after a routine extubation. Surprisingly, NIV not only failed to lower the reintubation rate compared to conventional therapy (approximately 50% in both groups) but it also increased ICU mortality. This was thought to be related to delayed reintubations in the NIV group, an average of 10 h later than in the conventional therapy group. The authors 51 concluded that NIV is not effective in averting the need for reintubation in unselected patients in whom respiratory failure develops after extubation and that it may in fact be harmful. However, only 10% of the enrolled patients had COPD. Also, approximately 25% of the conventional therapy group was crossed over to NIV when they met criteria for respiratory failure, and only 25% of these patients required reintubation. Thus, results of the study 51 do not apply to COPD patients and suggest that rather than initiating NIV early in patients deemed at risk for postextubation failure, one should wait until there are clear indications for NIV so that appropriate patients can be selected. Two subsequent trials support these latter inferences. Ferrer et al 52 identified patients at risk for postextubation failure by virtue of age 65 years, a history of CHF, or an APACHE (acute physiology and chronic health evaluation) II score 12. Although postextubation respiratory failure and need for intubation were significantly reduced by NIV overall, most of the benefit was attributable to the hypercapnic subgroup, amounting to about one third of the patients, who also had a significantly lower mortality rate than control subjects (4% vs 41%, p 0.05). Nava et al 53 used NIV in patients at high risk for extubation failure, using criteria similar to those of Esteban et al, 51 but risk was higher because patients had to have failed at least one T-piece trial. Once again, the need for reintubation (8% vs 18%, p 0.027) was reduced compared to the conventional therapy group. Also, ICU mortality was 10% less in the NIV group (p 0.01) mediated by the reduced need for reintubation. Thus, a trial of NIV appears to be warranted in patients at high risk for extubation failure, particularly if they have hypercapnic respiratory failure. Selection of Patients for NIV Selection of appropriate patients is crucial for the optimization of NIV success rates and resource utilization. Often, NIV must be started before laboratory data are available because patients may deteriorate during the delay and increase the risk of NIV failure. As depicted in Figure 1, patients with respiratory distress and an appropriate diagnosis should be considered for NIV. At the bedside, the clinician must make two fundamental judgments: (1) whether the patient needs ventilatory assistance based on symptoms and signs of increased work of breathing or arterial blood gas derangements, and (2) whether such patients are candidates for NIV or should be promptly intubated. These determinations are key to the appropriate application and outcome of NIV and are based on the diagnosis, bedside observations, the clinician s experience, and consideration of available guidelines (Table 1). The timing of NIV initiation is important too. NIV should be started early, as soon 714 Postgraduate Education Corner

Figure 1. Algorithm illustrating the principles of patient selection and practical application of NIV. Patients are started on NIV if respiratory distress develops in the setting of de novo or acute-on-chronic respiratory failure, or following surgery or extubation. They should have an appropriate diagnosis and meet guidelines demonstrating the need for ventilatory assistance and absence of contraindications. After starting NIV, they should be closely monitored and checked at 1 to 2 h to establish that they are responding favorably. If they have ALI/ARDS and are not good candidates, have contraindications or fail the 1- to 2-h checkpoint, they should be intubated unless they have a DNI status, in which case some patients might still benefit from palliation of respiratory distress. NIV to facilitate weaning should be considered for intubated patients. Patients who respond favorably to NIV should be monitored closely and reassessed periodically to determine whether they are ready to attempt weaning, which is usually accomplished by temporary discontinuation. If they have persisting respiratory failure after temporary discontinuation, long-term nocturnal NIV should be considered. www.chestjournal.org CHEST / 132 / 2/ AUGUST, 2007 715

Table 1 Selection Guidelines for NIV in the Acute Setting Appropriate diagnosis with potential reversibility Establish need for ventilatory assistance Moderate-to-severe respiratory distress and Tachypnea (respiratory rate 24/min for COPD, 30/min for CHF); accessory muscle use or abdominal paradox Blood gas derangement (ph 7.35, Paco 2 45 mm Hg, or Pao 2 /Fio 2 200) Exclude patients with contraindications to NIV Respiratory or cardiac arrest Medical instability (hypotensive shock, myocardial infarction requiring intervention, uncontrolled ischemia or arrhythmias) Unable to protect airway Unable to fit mask Untreated pneumothorax Recent upper airway or esophageal surgery Excessive secretions* Uncooperative or agitated* *Relative contraindications. as indications appear, because delay may permit further deterioration and increase the likelihood of failure. 54 Coma has been considered a contraindication to NIV in the past, but in a prospective cohort study, Gonzalez Diaz et al 55 observed a high success rate of NIV in patients with hypercapnic coma. Also, Scala et al 56 showed in a case study that NIV may be successfully used in COPD patients with acute respiratory failure and altered consciousness, although more severely impaired consciousness was associated with higher failure rates. Predictors of NIV success or failure may also be helpful in selecting patients (Table 2). The best predictor of success is a favorable response to NIV within the first 2 h. In a prospective cohort study of nearly 800 COPD patients treated with NIV, Confalonieri et al 57 identified four factors APACHE II score, ph, respiratory rate, and Glasgow coma score that, when combined in a chart, showed good predictive value at baseline. These factors had even better predictive value after 2hofNIVuse; if all four factors were favorable, the chance of success was 97%; whereas if all were unfavorable, failure was a virtual certainty (99%). Antonelli et al 47 made similar observations in patients with hypoxemic respiratory failure: if Pao 2 / Fio 2 failed to increase 146 after the first hour of NIV therapy, or if the patient had pneumonia and ARDS, the risk of NIV failure was increased. These observations, combined with those of Esteban et al, 51 demonstrating worse outcomes in NIV-treated patients having delayed reintubations, emphasize the importance of carefully reassessing patients soon after NIV initiation (1- to 2-h checkpoint as depicted in Fig 1). If they fail to improve sufficiently, they Table 2 Factors Associated With NIV Success in the Acute Setting Synchronous breathing with ventilator Dentate Less air leaking Fewer secretions Good tolerance Respiratory rate 30/min* Lower APACHE II score ( 29)* ph 7.30* Glasgow coma score 15* Pao 2 /Fio 2 146 after first hour if hypoxemic respiratory failure COPD, CPE No pneumonia, ARDS Best predictor of success is a good response to NPPV within 1 to 2h: Reduction in respiratory rate Improvement in ph Improvement in oxygenation Reduction in Paco 2 *If all four are present in COPD patients at baseline, the likelihood of success is 94%; if present after 2hoftherapy, the likelihood of success is 97%. 46 should be promptly intubated because a delay in needed intubation permits the development of a respiratory crisis, requiring emergent intubation and increasing the likelihood of morbidity or mortality. Interfaces Advances in Technology A well-fitting and comfortable interface (or mask) is crucial to the success of noninvasive ventilation. Although nasal masks have certain advantages over oronasal (or full face) masks including greater comfort, less likelihood of causing claustrophobia, and easier speech and expectoration, they also permit more air leakage through the mouth and have been associated with a higher rate of initial intolerance during acute applications of NIV. 58 Thus, oronasal masks are preferred initially for most critical care applications, although a nasal mask should still be considered for patients with claustrophobia or frequent expectoration or for long-term applications. Other mask types that are receiving attention include the Total Face Mask (Respironics; Murrysville, PA), which seals around the perimeter of the face and may enhance mask tolerance in some patients, and the helmet, which has not yet been approved by the Food and Drug Administration for NIV in the United States but has been investigated in Europe. 59,60 The latter device consists of a plastic cylinder that fits over the head and seals around the neck and shoulders. Compared to the full face mask in a case-control study 61 of NIV to treat COPD 716 Postgraduate Education Corner

patients with acute respiratory failure, the helmet achieved similar improvements in vital signs, equivalent intubation and mortality rates and caused fewer complications, but Paco 2 tended to be higher at the end of the treatment period despite a higher level of pressure support. Also, noise levels within the helmet may be as high as 100 decibels, compared to 70 decibels with a full-face mask. 62 Thus, although the helmet has some advantages over the full face mask with regard to comfort and complications, it has other disadvantages including less efficient CO 2 removal and noisiness that limit its current utility. Dead Space and Rebreathing By virtue of its single-ventilator-circuit design, bilevel ventilation has raised concerns about rebreathing. 63 A lung model study 64 demonstrated that masks with smaller volumes were associated with less rebreathing and an in-mask exhalation port minimized rebreathing compared to an in-line port. Another lung model study 65 used a mannequin face to demonstrate that an in-mask exhalation port over the bridge of the nose minimized dynamic dead space, sometimes to levels below physiologic, presumably by flushing CO 2 from the nose and mouth. The correlation between dynamic dead space and actual mask volume was poor, probably because of air streaming, and dead space was also minimized if positive expiratory pressure flushed CO 2 from the ventilator tubing. Whether these differences in dead space and rebreathing are clinically important remains unclear, but these studies support the use of in-mask exhalation ports and positive expiratory pressure during bilevel ventilation. Ventilators NIV is usually delivered either by blower-based portable positive pressure bilevel ventilators derived from home-based CPAP systems or critical care ventilators designed to deliver invasive mechanical ventilation. No study has shown better NIV success rates for one type of ventilator than the other, but the ventilator mode used and specific settings are important for patient comfort and decreased work of breathing. Pressure support ventilation is rated as more tolerable by patients than assist-control modes, 66 and some studies 67 69 have demonstrated greater comfort with proportional assist ventilation than with pressure support, presumably because it is targeted to inspiratory flow as a surrogate of patient effort and can respond nearly instantaneously to changes in demand. Proportionalassist ventilation has only recently been approved by the Food and Drug Administration, but it has been available elsewhere in the world for almost a decade. The perceived need for multiple adjustments to compensate for patient elastance and resistance as well as added cost have probably limited greater use of this mode despite the finding in one of the controlled trials that proportional assist required fewer adjustments than pressure support. 68 Other desirable attributes of ventilators for NIV include the ability to compensate for air leaks, which helps to assure delivery of adequate tidal volumes. 70 Because NIV lowers humidity of delivered gas, humidification is useful to bring relative humidity back toward the ambient range, possibly enhancing comfort. 71 Heated passover humidifiers have minimal effects on delivered pressures, whereas heat and moisture exchangers are to be avoided because they can interfere with the ability of NIV to reduce work of breathing. 72 Ventilator Settings L Her et al 73 showed that patients with acute lung injury treated with NIV require pressure support levels of at least 10 to 15 cm H 2 O to reduce work of breathing. Not surprisingly, higher levels of positive end-expiratory pressure (PEEP) [10 cm H 2 Ovs5cm H 2 O] were more effective at improving oxygenation. Combining higher levels of pressure support with high-level PEEP can detract from patient comfort, however, so compromises may be necessary to optimize settings; maximal oxygenation may have to be sacrificed if patient comfort and reduction in work of breathing are prioritized. Many noninvasive ventilators now offer adjustable rise times or pressurization rates the time taken to achieve the target inspiratory pressure to optimize patient comfort. Prinianakis et al 74 found that a rapid pressurization rate was most effective at reducing work of breathing in COPD patients, but a slightly slower rate was associated with better comfort ratings. Guidelines, Utilization, and Outcomes Sinuff et al 75 found that a NIV guideline influenced caregiver behavior, leading to greater ICU utilization and more ordering of pulmonary consultations and arterial blood gases. However, overall mortality rate was unchanged and, of concern, the mortality rate actually increased in patients without COPD or CHF as the cause of their acute respiratory failure, who were excluded from NIV use by the guideline. The results highlight the need for ongoing guideline evaluation and modification because they could increase resource utilization and the cost of care if they mandate ICU use and frequent laboratory testing among all NIV patients, some of whom could conceivably be managed in less costly environments. www.chestjournal.org CHEST / 132 / 2/ AUGUST, 2007 717

Studies of NIV utilization in the acute care setting have found that enormous disparity exists between different institutions. In a 1997 survey of NIV use in European ICUs, Carlucci et al 76 found that 20% of ICUs surveyed used no NIV at all and, overall, NIV was used in 16% of all ventilator starts. A subsequent UK survey 77 found that 52% of hospitals were not using NIV. A more recent survey 78 in the United States found that although only 1 of 71 responding hospitals used no NIV, some used it in 5% of ventilator starts and others in 50%. Overall average use among ventilator starts was 20%, but only a third of patients with COPD or CHF received NIV as their initial ventilator therapy. Major reasons for not using NIV more were lack of physician knowledge and inadequately trained staff, suggesting that education may help to enhance utilization. However, progress is being made, as indicated by Demoule et al, 79 who found that the overall percentage for NIV among ventilator starts in European ICUs had risen to 23% by 2002; and Girou et al, 80 who showed that increasing use of NIV in CHF and COPD patients (from approximately 20 to 90% of ventilator starts) over a 7-year period in a French ICU was associated with a decrease in the rate of nosocomial pneumonias from 20 to 8% and in ICU mortality rate from 21 to 7% (all p 0.05). The latter study 80 also illustrates the value of increasing experience using NIV after establishing an NIV program. Conclusion The role of NIV in the management of acute respiratory failure has been further clarified in recent years. Evidence is strong to support the use of NIV in the initial management of acute respiratory failure in patients with COPD exacerbations, acute CPE, and immunocompromised states, and to facilitate extubation in patients with COPD with failed spontaneous breathing trials. A trial of NIV is justified in patients with asthma exacerbations, postoperative respiratory failure, extubation failure, hypoxemic respiratory failure or a DNI status, but because supporting evidence is not as strong, they should be carefully selected according to available guidelines and clinical judgment, taking into account risk factors for NIV failure. Once begun, patients should be closely monitored in an ICU or step-down unit until adequately stabilized, paying attention not only to vital signs and gas exchange, but also to comfort and tolerance. If patients do not have a favorable initial response to NIV, clinicians should strongly consider intubation without delay. When used appropriately, NIV improves patient outcomes and the efficiency of care. Although it is still used in only a select minority of patients with acute respiratory failure, it has assumed an important role in the therapeutic armamentarium. With technical advances and new evidence on its proper application, this role is likely to expand. References 1 Liesching T, Kwok H, Hill NS. Acute applications of noninvasive positive pressure ventilation. Chest 2003; 124:699 713 2 Brochard L, Isabey D, Piquet J, et al. Reversal of acute exacerbations of chronic obstructive lung disease by inspiratory assistance with a face mask. N Engl J Med 1990; 323:1523 1530 3 Bott J, Carroll MP, Conway JH, et al. Randomised controlled trial of nasal ventilation in acute ventilatory failure due to chronic obstructive airways disease. Lancet 1993; 341:1555 1557 4 Kramer N, Meyer TJ, Meharg J, et al. Randomized, prospective trial of noninvasive positive pressure ventilation in acute respiratory failure. Am J Respir Crit Care Med 1995; 151: 1799 1806 5 Plant PK, Owen JL, Elliott MW. Early use of non-invasive ventilation for acute exacerbations of chronic obstructive pulmonary disease on general respiratory wards: a multicentre randomised controlled trial. Lancet 2000; 355:1931 1935 6 Dikensoy O, Ikidag B, Filiz A, et al. Comparison of noninvasive ventilation and standard medical therapy in acute hypercapnic respiratory failure: a randomised controlled study at a tertiary health centre in SE Turkey. Int J Clin Pract 2002; 56:85 88 7 Celikel T, Sungur M, Ceyhan B, et al. Comparison of noninvasive positive pressure ventilation with standard medical therapy in hypercapnic acute respiratory failure. Chest 1998; 114:1636 642 8 Ram FS, Picot J, Lightowler J, et al. Non-invasive positive pressure ventilation for treatment of respiratory failure due to exacerbations of chronic obstructive pulmonary disease. Cochrane Database Syst Rev 2004; 3:CD004104 9 Keenan SP, Sinuff T, Cook DJ, et al. Which patients with acute exacerbation of chronic obstructive pulmonary disease benefit from noninvasive positive pressure ventilation? A systematic review of the literature. Ann Intern Med 2003; 138:861 870 10 Bersten AD, Holt AW, Vedig AE, et al. Treatment of severe cardiogenic pulmonary edema with continuous positive airway pressure delivered by face mask. N Engl J Med 1991; 325:1825 1830 11 Lin M, Yang Y, Chiany H, et al. Reappraisal of continuous positive airway pressure therapy in acute cardiogenic pulmonary edema: short-term results and long-term follow-up. Chest 1995; 107:1379 1386 12 Nava S, Carbone G, DiBattista N, et al. Noninvasive ventilation in cardiogenic pulmonary edema: a multicenter randomized trial. Am J Respir Crit Care Med 2003; 168:1432 1437 13 Crane SD, Elliott MW, Gilligan P, et al. Randomised controlled comparison of continuous positive airways pressure, bilevel non-invasive ventilation, and standard treatment in emergency department patients with acute cardiogenic pulmonary oedema. Emerg Med J 2004; 21:155 161 14 Pang D, Keenan SP, Cook DJ, et al. The effect of positive airway pressure on mortality and the need for intubation in cardiogenic pulmonary edema. Chest 1998; 114:1185 1192 15 Rasanen J, Heikkila J, Downs J, et al. Continuous positive airway pressure by face mask in acute cardiogenic pulmonary 718 Postgraduate Education Corner

edema. Am J Cardiol 1985; 55:296 300 16 Lin M, Chiang HT. The efficacy of early continuous positive airway pressure therapy in patients with acute cardiogenic pulmonary edema. J Formos Med Assoc 1991; 90:736 743 17 Park M, Sangean MC, Volpe Mde S, et al. Randomized, prospective trial of oxygen, continuous positive airway pressure, and bilevel positive airway pressure by face mask in acute cardiogenic pulmonary edema. Crit Care Med 2004; 32:2407 2415 18 Masip J, Roque M, Sanchez B, et al. Noninvasive ventilation in acute cardiogenic pulmonary edema. JAMA 2005; 294: 3124 3130 19 Winck JC, Azevedo LF, Costa-Pereira A, et al. Efficacy and safety of non-invasive ventilation in the treatment of acute cardiogenic pulmonary edema: a systematic review and metaanalysis. Crit Care 2006; 10:R69 20 Collins SP, Mielniczuk LM, Whittingham HA, et al. The use of noninvasive ventilation in emergency department patients with acute cardiogenic pulmonary edema: a systematic review. Ann Emerg Med 2006; 48:260 269 21 Ho KM, Wong K. A comparison of continuous and bi-level positive airway pressure non-invasive ventilation in patients with acute cardiogenic pulmonary oedema: a meta-analysis. Crit Care 2006; 10:R4921 22 Mehta S, Jay GD, Woolard RH, et al. Randomized prospective trial of bilevel versus continuous positive airway pressure in acute pulmonary edema. Crit Care Med 1997; 25:620 628 23 Ferrer M, Esquinas A, Arancibia F, et al. Noninvasive ventilation during persistent weaning failure: a randomized controlled trial. Am J Respir Crit Care Med 2003; 168:70 76 24 Nava S, Ambrosino N, Clini E, et al. Non-invasive mechanical ventilation in the weaning of patients with respiratory failure due to chronic obstructive pulmonary disease: a randomized study. Ann Intern Med 1998; 128:721 728 25 Antonelli M, Conti G, Bufi M, et al. Noninvasive ventilation for treatment of acute respiratory failure in patients undergoing solid organ transplantation: a randomized trial. JAMA 2000; 283:2239 2240 26 Hilbert G, Gruson D, Vargas F, et al. Noninvasive ventilation in immunosuppressed patients with pulmonary infiltrates, and acute respiratory failure. N Engl J Med 2001; 344:481 487 27 Meduri GU, Turner RE, Abou-Shala N, et al. Noninvasive positive pressure ventilation via face mask: first line intervention in patients with acute hypercapnic and hypoxemic respiratory failure. Chest 1996; 109:179 193 28 Meduri GU, Cook TR, Turner RE, et al. Noninvasive positive pressure ventilation in status asthmaticus. Chest 1996; 110: 767 774 29 Fernandez MM, Villagra A, Blanch L, et al. Non-invasive mechanical ventilation in status asthmaticus. Intensive Care Med 2001; 27:486 492 30 Soroksky A, Stav D, Shpirer I. A pilot prospective, randomized, placebo-controlled trial of bi-level positive airway pressure in acute asthmatic attack. Chest 2003; 123:1018 1025 31 Agarwal R, Malhotra P, Gupta D. Failure of NIV in acute asthma: case report and a word of caution. Emerg Med J 2006; 23:e9 32 Ram FS, Wellington S, Rowe B, et al. Non-invasive positive pressure ventilation for treatment of respiratory failure due to severe acute exacerbations of asthma. Cochrane Database Syst Rev 2005; 1:CD004360 33 Kindgen-Milles D, Muller E, Buhl R, et al. Nasal continuous positive airway pressure reduces pulmonary morbidity and length of stay following thoracoabdominal aortic surgery. Chest 2005; 128:821 828 34 Squadrone V, Coha M, Cerutti E, et al. Continuous positive airway pressure for treatment of postoperative hypoxemia. JAMA 2005; 293:589 595 35 Auriant I, Jallot A, Herve P, et al. Noninvasive ventilation reduces mortality in acute respiratory failure following lung resection. Am J Respir Crit Care Med 2001; 164:1231 1235 36 Clarke DE, Vaughan L, Raffin TA. Noninvasive positive pressure ventilation for patients with terminal respiratory failure: the ethical and economic cost of delaying the inevitable are too great. Am J Crit Care 1994; 3:4 5 37 Levy MM, Tanios MA, Nelson D, et al. Outcomes of patients with do-not- intubate orders treated with noninvasive ventilation. Crit Care Med 2004; 32:2002 2007 38 Schettino G, Altobelli N, Kacmarek RM. Noninvasive positive-pressure ventilation reverses acute respiratory failure in select do-not-intubate patients. Crit Care Med. 2005; 33: 1976 1982 39 Curtis RJ, Cook DJ, Sinuff T, et al. Noninvasive positive pressure ventilation in critical and palliative care settings: understanding the goals of therapy. Crit Care Med 2007 (in press) 40 Antonelli M, Conti G, Rocco M, et al. A comparison of noninvasive positive-pressure ventilation and conventional mechanical ventilation in patients with acute respiratory failure. N Engl J Med 1998; 339:429 435 41 Ferrer M, Esquinas A, Leon M, et al. Noninvasive ventilation in severe hypoxemic respiratory failure: a randomized clinical trial. Am J Respir Crit Care Med 2003; 168:1438 1444 42 Ambrosino N, Foglio K, Rubini F, et al. Noninvasive mechanical ventilation in acute respiratory failure due to chronic obstructive pulmonary disease: correlates for success. Thorax 1995; 50:755 757 43 Honrubia T, Garcia Lopez F, Franco N, et al. Noninvasive vs. conventional mechanical ventilation for acute respiratory failure. Chest 2005; 128:3916 3924 44 Antonelli M, Conti G, Moro ML, et al. Predictors of failures of noninvasive positive pressure ventilation in patients with acute hypoxemic respiratory failure: a multi-center study. Intensive Care Med 2001; 27:1718 1728 45 Jolliet P, Abajo B, Pasquina P, et al. Non-invasive pressure support ventilation in severe community-acquired pneumonia. Intensive Care Med 2001; 27:812 821 46 Confalonieri M, Potena A, Carbone G, et al. Acute respiratory failure in patients with severe community-acquired pneumonia. Am J Respir Crit Care Med 1999; 160:1585 1591 47 Antonelli M, Conti G, Esquinas A, et al. A multiple-center survey on the use in clinical practice of noninvasive ventilation as a first-line intervention for acute respiratory distress syndrome. Crit Care Med 2007; 35:18 25 48 Rana S, Hussam J, Gay P, et al. Failure of non-invasive ventilation in patients with acute lung injury: observational cohort study. Crit Care 2006; 10:R79 49 Blivet S, Philit F, Sab JM, et al. Outcome of patients with idiopathic pulmonary fibrosis admitted to the ICU for respiratory failure. Chest 2001; 120:8 10 50 Nevins ML, Epstein SK. Predictors of outcome for patients with COPD requiring invasive mechanical ventilation. Chest 2001; 119:1840 1849 51 Esteban A, Frutos-Vivar F, Ferguson ND, et al. Noninvasive positive-pressure ventilation for respiratory failure after extubation. N Engl J Med 2004; 350:2452 2460 52 Ferrer M, Valencia M, Nicolas JM, et al. Early noninvasive ventilation averts extubation failure in patients at high risk: a randomized trial. Am J Respir Crit Care Med 2006; 173:164 170 53 Nava S, Gregoretti C, Fanfulla F, et al. Noninvasive ventilation to prevent respiratory failure after extubation in high-risk patients. Crit Care Med 2005; 33:2465 470 www.chestjournal.org CHEST / 132 / 2/ AUGUST, 2007 719

54 Nava S, Navalesi P, Conti G. Time of non-invasive ventilation. Intensive Care Med 2006; 32:361 370 55 Gonzalez Diaz G, Carillo A, Perez P, et al. Noninvasive positive-pressure ventilation to treat hypercapnic coma secondary to respiratory failure. Chest 2005; 127:952 960 56 Scala R, Naldi M, Archinucci I, et al. Noninvasive positivepressure ventilation in patients with acute exacerbations of COPD and varying levels of consciousness. Chest 2005; 128:1657 1666 57 Confalonieri M, Garuti G, Cattaruzza MS, et al. A chart of failure risk for noninvasive ventilation in patients with COPD exacerbation. Eur Respir J 2005; 25:348 355 58 Kwok H, McCormack J, Cece R, et al. Controlled trial of oronasal versus nasal mask ventilation in the treatment of acute respiratory failure. Crit Care Med 2003; 31:468 473 59 Tonnelier JM, Prat G, Nowak E, et al. Noninvasive continuous positive airway pressure ventilation using a new helmet interface: a case prospective pilot study. Intensive Care Med 2003; 29:2077 2080 60 Principi T, Pantanetti S, Catani F, et al. Noninvasive continuous positive airway pressure delivered by helmet in hematological malignancy patients with hypoxemic acute respiratory failure. Intensive Care Med 2004; 30:147 150 61 Antonelli M, Pennisi MA, Pelosi P, et al. Noninvasive positive pressure ventilation using a helmet in patients with acute exacerbation of chronic obstructive pulmonary disease. Anesthesiology 2004; 100:16 24 62 Cavaliere F, Conti G, Costa R, et al. Noise exposure during noninvasive ventilation with a helmet, a nasal mask, and a facial mask. Intensive Care Med 2004; 30:1755 1760 63 Ferguson GT, Gilmartin M. CO 2 rebreathing during BiPAP ventilatory assistance. Am J Respir Crit Care Med 1995; 151:1126 1135 64 Schettino GPP, Chatmongkolchart S, Hess D, et al. Position of exhalation port and mask design affect CO 2 rebreathing during noninvasive positive pressure ventilation. Crit Care Med 2003; 31:2178 2182 65 Saatci E, Miller DM, Sztell IM, et al. Dynamic dead space in face masks used with noninvasive ventilators: a lung model study. Eur Respir J 2004; 23:129 135 66 Vitacca M, Rubini F, Foglio K, et al. Noninvasive modalities of positive pressure ventilation improve the outcome of acute exacerbations of COLD patients. Intensive Care Med 1993; 19:450 455 67 Fernandez-Vivas M, Caturia-Such J, de la Rosa JG, et al. Noninvasive pressure support versus proportional assist ventilation in acute respiratory failure. Intensive Care Med 2003; 29:1126 1133 68 Gay PC, Hess DR, Hill NS. Noninvasive proportional assist ventilation for acute respiratory insufficiency: comparison with pressure support ventilation. Am J Respir Crit Care Med 2001; 164:1606 1611 69 Wysocki M, Richard JC, Meshaka P. Noninvasive proportional assist ventilation compared with noninvasive pressure support ventilation in hypercapnic acute respiratory failure. Crit Care Med 2002; 30:323 329 70 Mehta S, McCool FD, Hill NS. Leak compensation in positive pressure ventilators: a lung model study. Eur Respir J 2001; 17:259 267 71 Holland AE, Denehy L, Buchan CA, et al. Efficacy of a heated passover humidifier during noninvasive ventilation: a bench study. Respir Care 2007; 52:38 44 72 Lellouche F, Maggiore SM, Deye N, et al. Effect of the humidification device on the work of breathing during noninvasive ventilation. Intensive Care Med 2002; 28:1582 1589 73 L Her E, Deye N, Lellouche F, et al. Physiologic effects of noninvasive ventilation during acute lung injury. Am J Crit Care Med 2005; 172:1112 1118 74 Prinianakis G, Delmastro M, Carlucci A, et al. Effect of varying the pressurization rate during noninvasive pressure support ventilation. Eur Respir J 2004; 23:314 320 75 Sinuff T, Cook DJ, Randall J, et al. Evaluation of a practice guideline for noninvasive positive pressure ventilation for acute respiratory failure. Chest 2003; 123:2062 2073 76 Carlucci A, Richard JC, Wysocki M, et al. Noninvasive versus conventional mechanical ventilation. Am J Respir Crit Care Med 2001; 163:874 880 77 Doherty MJ, Greenstone MA. Survey of non-invasive ventilation (NIPPV) in patients with acute exacerbations of chronic obstructive pulmonary disease (COPD) in the UK. Thorax 1998; 53:863 866 78 Maheshwari V, Paioli D, Rothaar R, et al. Utilization of noninvasive ventilation in acute care hospitals. Chest 2006; 129:1226 1233 79 Demoule A, Girou E, Richard JC, et al. Increased use of noninvasive ventilation in French intensive care units. Intensive Care Med 2006; 32:1747 1755 80 Girou E, Brun-Buisson C, Taille S, et al. Secular trends in nosocomial infections and mortality associated with noninvasive ventilation in patients with exacerbation of COPD and pulmonary edema. JAMA 2003; 290:2985 2991 720 Postgraduate Education Corner