Chronic obstructive pulmonary disease is associated with important morbidity–mortality and affects up to 10% of the Spanish population according to the EPI-SCAN study.6 The exacerbation of COPD is the cause of 20% of all emergency care consultations, and it is estimated that about 25% of the affected patients present acidosis upon arrival. Management of the exacerbations is based on drug treatment and ventilatory support. Evidence of the benefits of NIV in hypercapnic ARF among patients with COPD has been widely documented in the scientific literature, and the technique is warranted both by a high degree of recommendation in the different clinical guides7 and by the progressive increase in its use by specialists in recent years. A registry of hospitals in the United States during the period 1998–2008 referred to the use of MV in patients with COPD exacerbation revealed a progressive 462% increase in the use of NIV over the years (from 1% to 4.5% of all admissions), together with a 42% decrease in the use of invasive MV (from 6% to 3.5% of all admissions).8 In Europe, during the year 2008 a study on the use of NIV found the management of hypercapnic ARF to be the main indication of the technique (48% of the cases).9

Chronic obstructive pulmonary disease exacerbation is characterized by an increase in mechanical loading due to the high airflow resistances and dynamic hyperinsufflation which initially give rise to an increase in respiratory effort and intrinsic PEEP. In later stages the condition evolves toward muscle fatigue and consequent clinical and blood gas deterioration. Dynamic hyperinsufflation occurs because of an expiratory time that is too short to reach the resting volume of the respiratory system, and increases with each successive respiration. Under these circumstances, NIPPV has beneficial effects, which will be summarized below and are represented in Fig. 1. On one hand, application of a positive inspiratory pressure reduces respiratory labor with an increase in alveolar ventilation and a lowering of the respiratory frequency–this in turn produces a prolongation of the expiratory time with lesser air trapping. In relation to gas exchange, and due to the increase in minute volume, we observe a lowering of PaCO2 and an increase in pH. On the other hand, the application of external PEEP counters the inspiratory effort needed to surpass the intrinsic PEEP due to the dynamic hyperinsufflation and which can originate up to 60% of the increase in respiratory effort–thereby reducing muscle work. Fig. 2 proposes an interventional algorithm for patients presenting COPD with hypercapnic ARF.

Figure 1.

Physiopathology of hypercapnic acute respiratory failure in COPD and the processes that are blocked by the action of mechanical ventilation (represented in black). RF, respiratory frequency; PEEPext, positive end-expiratory pressure programmed by the ventilator; PEEPi, intrinsic PEEP; Texp, expiratory time; MV, mechanical ventilation; NIPPV, noninvasive positive pressure ventilation.

(0.19MB).

Figure 2.

Application of noninvasive ventilation in acute respiratory failure in chronic obstructive pulmonary disease with hypercapnic decompensation, and in acute cardiogenic pulmonary edema. The ventilatory parameters used are orientative according to tolerance of the technique and the presence of leakage. CPAP, continuous positive airway pressure; ACPE, acute cardiogenic pulmonary edema; COPD, chronic obstructive pulmonary disease; RF, respiratory frequency; PEEP, positive end-expiratory pressure; NIPPV, noninvasive ventilation with inspiratory support (PS, pressure support; BiPAP, proportional assist ventilation, etc.); TV, tidal volume.

(0.25MB).

Hypercapnic ARF in patients with COPD is the main indication of NIPPV–the first studies having been published over two decades ago.10

Brochard et al.10 evaluated the physiological benefits of NIPPV in patients with exacerbated COPD and observed significant improvement of dyspnea, respiratory frequency and gas exchange, with favorable changes in pH, PaCO2 and PaO2. After comparing the results with an oxygen therapy control group, the use of NIPPV was seen to be associated with a lesser need for intubation (7% vs 77%, p<0.001) and a shorter stay in the ICU (7 vs 19days, p<0.01).

Posteriorly, the same group conducted a randomized study of 85 patients with exacerbated COPD,11 comparing the use of NIPPV with conventional oxygen therapy. The NIPPV group showed significant improvement in blood gases (pH, PaCO2 and PaO2) and respiratory frequency, together with a significant decrease in the need for intubation (26% vs 74%, p=0.001), the presence of complications, hospital stay (23 vs 35days, p=0.02) and hospital mortality (9% vs 29%, p=0.02).

Since then, many studies have confirmed the benefits of NIPPV in patients with COPD and hypercapnic ARF in terms of a lesser need for intubation, diminished hospital mortality and a shortened hospital stay.12 Different metaanalyses have examined these results, revealing significant decreases in the need for intubation (18–28%) and in hospital mortality (10–13%),13,14 together with a significant shortening of hospital stay (2–5days)14–17 (Table 1).

Most of the patients included in the abovementioned studies had respiratory acidosis. On considering the studies that included patients with less severe exacerbations (initial pH>7.30), NIPPV did not show advantages with respect to conventional oxygen therapy,18,19 as has been confirmed in a metaanalysis13 on evaluating the subgroup of patients with pH>7.30. In contrast, another metaanalysis that did not include the study of Keenan et al.18 detected a significant reduction in the percentage of intubations and in mortality in this subgroup of patients15 (Table 1).

In sum, the application of NIPPV together with regular medical treatment in patients with COPD and hypercapnic ARF reduces the incidence of orotracheal intubation, hospital mortality and hospital stay.

Invasive mechanical ventilation for over three days in patients with COPD is associated with complications such as muscle atrophy and an increased risk of nosocomial pneumonia. It therefore seems prudent to design strategies to shorten the duration of MV. On the contrary, nosocomial pneumonia is associated with increased patient mortality. NIPPV could be used as ventilatory support after invasive MV in three contexts: to facilitate weaning from invasive MV, to prevent postextubation ARF, or for the management of postextubation ARF.

The pressure support mode has commonly been used in NIV applied to hypercapnic ARF in COPD patients. Since a constant level of inspiratory support is provided, a decrease is observed in the variability of the respiratory pattern, and this in turn may be associated with asynchronies and overdistension. More recent modes have been introduced that use a more physiological inspiratory support system in order to improve synchronization between the patient and the ventilator. One of these modes is proportional assist ventilation (PAV), in which inspiratory support proportional to the inspiratory flow is provided. The degree of support is programmed according to the resistance of the airway and compliance. However, although this mode is interesting, it has not yet been able to demonstrate advantages with respect to pressure support.29 Another recently introduced ventilation mode is the neurally adjusted ventilatory assist mode (NAVA), which is based on the detection of diaphragm contraction through esophageal electrodes. NIV with this mode appears to reduce the severe asynchronies observed with pressure support, while affording similar results in terms of gas exchange.30

The percentage of failure of NIPPV in patients with COPD and hypercapnic ARF that finally need endotracheal intubation is less than 25%.11,12,31,32 The parameters associated with success and failure of the technique are summarized in Table 3. An adequate level of patient consciousness at the start of NIPPV, and the improvement of pH, PaCO2 and level of consciousness after one hour of NIPPV are related to a good response to NIPPV in hypercapnic ARF in patients with COPD.32 In patients in whom NIV fails, a recent observational study indicates that the range of time on NIV prior to intubation is very wide and that an increase in the duration of NIV prior to intubation is associated with a decrease in hospital survival–though only in the group of patients without previous cardiac or respiratory disease33–hence the importance of the time window for evaluating the success or failure of the technique, with a view to avoiding delays in intubation that could worsen the prognosis.

On the other hand, earliness in the introduction of NIPPV is fundamental for the success of the technique. Patient extenuation, which is associated with intense respiratory acidosis and possible encephalopathy, is to be avoided since these conditions are characterized by a high failure risk. Likewise, the presence of pneumonia or acute respiratory distress syndrome (ARDS) is also associated with failure of the technique.34

Poor tolerance of the technique is associated with NIPPV failure–asynchrony being one of the most relevant factors in this sense. Asynchrony is defined by the presence of an ineffective trigger, double trigger, auto-trigger, premature cycling or late cycling, and is present in up to 43% of all patients–the level of pressure support and the magnitude of leakage being the main factors responsible for severe asynchronies.35

The type of mask used also plays an important role in defining patient tolerance of the technique. The current main interfaces comprise nasal, oronasal, facial or helmet masks. In this context, oronasal masks are the preferred option in the presence of ARF for both patients and specialists.9Table 4 describes the main advantages and inconveniences of the different interfaces.

Although most studies on NIPPV have been carried out in ICUs, the use of the technique has spread in recent years to Emergency Care Departments and Pneumology wards, with equally satisfactory results.15 In a European registry published in 2008,9 the use of NIPPV was found to be more widespread among pneumologists than among intensivists or anesthetists. In patients with NIV failure risk (Table 3), the technique should be used in units that have expert professionals and are able to react quickly in the event of failure of the technique.31,36 These requirements are met by Intensive or Semicritical Care Units. When NIPPV is started outside this setting and gas exchange, respiratory frequency or dyspnea fails to improve; the degree of encephalopathy worsens; or hemodynamic deterioration is moreover observed, then transfer of these patients to such units should be considered.

Heart failure affects 2–3% of the population, and its prevalence progressively increases with age–reaching 10–20% among patients between 70 and 80 years of age. The mean age of patients with heart failure in the industrialized world is 75 years. The disease is the cause of 5% of all emergency hospital admissions and occupies 10% of the global hospital beds.37 On the other hand, NIV is used in 80% of the episodes of acute cardiogenic pulmonary edema (ACPE).31

Acute cardiogenic pulmonary edema is characterized by an increase in left ventricle filling pressures that retrogradely generate a rise in pulmonary capillary pressure, followed by fluid extravasation toward the pulmonary interstitial compartment and alveolar spaces. All these in turn produce an increase in airway resistances, a decrease in lung diffusion capacity, a drop in functional residual capacity, and an increased intrapulmonary shunt effect. The result is the development of hypoxemia with an increase in respiratory effort.

The application of intrathoracic positive pressure in patients with lung edema, particularly with PEEP, whether with or without inspiratory support, on one hand produces a decrease in venous return and in right ventricle preload, and on the other reduces left ventricle postload secondary to a decrease in transmural pressure (transmural pressure=intraventricular pressure−intrathoracic pressure), thereby improving cardiac output and myocardial contractility. Furthermore, in these patients, the application of positive pressure favors alveolar recruitment and increases functional residual capacity, lung compliance and alveolar ventilation, with a reduction of the intrapulmonary shunt and respiratory effort, and improved oxygenation (Table 5).

Separate mention is required of situations of cardiogenic shock or right ventricular failure, in which the application of NIV may prove deleterious.37

Faster clinical (reduction of respiratory frequency and of dyspneic sensation) and blood gas improvements (increased PaO2, reduction of PaCO2 and of acidosis) have been observed with the application of NIV compared with conventional oxygen therapy,38 with both CPAP39,40 and NIPPV41,42–the differences being manifest from 30min of application of the technique.

Both NIPPV and CPAP show rapid response times (from 30min)43–clinical resolution generally being faster with the application of NIPPV,44,45 reaching differences of almost one hour between the two techniques.44 Likewise, studies with a physiological basis have shown that although the use of CPAP is associated with a decrease in respiratory effort among patients with ACPE, this effect is greater when NIPPV is applied.46

Other studies report that NIV in ACPE may be especially beneficial for patients with hypercapnia, by reducing the percentage of intubations with respect to non-hypercapnic individuals.41,47 However, in the study published by Nava et al.,41 72% of the subgroup of hypercapnic patients consisted of individuals who moreover had COPD, while in the non-hypercapnic subgroup these individuals represented only 11%. This suggests that it is probably the underlying disease, not hypercapnia alone, that constitutes the marker of success. Nouira et al.44 conducted a sub-analysis of the patients with hypercapnia. In these subjects the authors compared CPAP vs NIPPV, and no differences in patient course were seen between the two techniques.

The mean duration of NIV in ACPE is between 2 and 4h.48 Hospital stay is not modified by the use of NIV vs conventional treatment, and is reported to be 8–14 days.38,39,41,42,48

The percentage of endotracheal intubation in patients with ACPE subjected to conventional oxygen therapy is 10–28%, depending on the series. However, when using NIV, this percentage drops to 3–12%, with both CPAP and NIPPV.49

The largest study to date on the effects of NIV in ACPE was published by Gray et al. in 2008.38 This study randomized 1069 patients to three treatment groups: CPAP, NIPPV and a control group administered conventional oxygen therapy. The authors observed faster improvement of dyspneic sensation and of pH with the use of CPAP or NIPPV in comparison with conventional oxygen therapy, with no significant differences in mortality or in the need for intubation between the groups. Likewise, no differences were recorded in hospital stay or in the incidence of myocardial infarction. It must be noted that this study excluded the more seriously ill patients (which are the subjects who possibly could derive greater benefit from the technique), with a global percentage of intubations of under 3%, and that moreover 15% of the patients in the conventional oxygen therapy group were rescued with NIV.

Most randomized studies report a clear tendency toward a lesser need for intubation with NIV.40,48,50,51 The different metaanalyses49,52–57 coincide in reporting a significant decrease in the percentage of intubations with the application of NIV, in comparison with conventional oxygen therapy (as can be seen in Table 6), on considering both CPAP and NIPPV–with calculation of an absolute reduction of the need for intubation of 22% and 18%, respectively.54

On separately analyzing the effect of the application of NIPPV or CPAP vs conventional oxygen therapy, both techniques have demonstrated their efficacy in reducing the need for intubation, particularly when CPAP is used. The global studies published to date have been unable to demonstrate the superiority of one technique over the other (Table 6), and in this sense similar intubation percentages are seen with the use of both CPAP and NIPPV in application to ACPE.

There is controversy regarding the impact of NIV upon hospital mortality in patients with ACPE. Although most randomized studies report a tendency toward lesser mortality in the groups treated with NIV, they have been unable to demonstrate significant reductions in patient mortality,38,48,51,58 probably because of the limited statistical power involved. The mortality rate among the patients receiving conventional oxygen therapy is about 16%, vs 10% in those who receive NIV.49

The effect of NIV upon hospital mortality has been reflected in different metaanalyses comparing the technique with conventional oxygen therapy.49,52–57 Although these analyses are based on studies with a limited number of patients and with variable methodological designs, they agree that the use of NIV is associated with a significant decrease in hospital mortality (Table 6).

On separately analyzing the two ventilation modes, the application of CPAP is seen to maintain a significant reduction in mortality compared with conventional treatment, with an absolute reduction of 13%,54 while the application of NIPPV only reveals a tendency toward decreased mortality (absolute reduction 7%).54 Nevertheless, no significant differences are observed on comparing the two techniques (Table 6).

The concern about acute myocardial infarction (AMI) during the application of NIV is mainly attributable to a randomized study involving patients with ACPE in which a comparison was made of NIPPV vs CPAP–a tendency toward more frequent AMI being observed in the group of patients treated with NIPPV. However, this study only included a total of 27 patients, and the NIPPV treatment group had a greater proportion of patients with chest pain and left bundle block at the start of the study.45 Posteriorly, numerous randomized studies have been unable to confirm these results,41,43,44,48,58,59 though it is true that most of them excluded patients with acute coronary syndrome (ACS) as a cause of ACPE43,44,59–particularly those patients amenable to immediate revascularization.41,48,58

The different metaanalyses likewise have not detected a relationship between AMI and the use of NIV in patients with ACPE49,53–55,57 either on comparing NIV with conventional oxygen therapy or on comparing the two NIV techniques with each other, both during treatment with NIV49,53,54,57 and after its application57 (Table 6).

On the basis of the existing results, the recommendation is to use NIV in ACPE in the absence of shock or ACS amenable to urgent revascularization, and making no distinction between CPAP and NIPPV.7

Treatment in ACPE must be provided on an early basis. This therefore also applies to NIV, which has been shown to be cost-effective provided it is introduced early. In the pre-hospital setting, a randomized study of 71 patients with ARF (67% of whom had ACPE) compared the use of conventional oxygen therapy vs CPAP (10cmH2O). A significant reduction was observed in the percentage of intubations and in patient mortality in the group treated with CPAP.60 In another study in the pre-hospital setting involving 124 patients with ACPE, a comparison was made of the effects of the immediate application of CPAP during the first 15min, followed by the addition of conventional treatment (oxygen and medication), vs the immediate provision of conventional treatment followed 15min later by the addition of CPAP.50 The early CPAP group showed faster improvement of the physiological parameters (dyspnea, respiratory frequency, PaCO2 and PaO2), a lesser need for inotropic drugs and intubation, and a tendency toward lesser mortality.

In applying CPAP, we usually employ CPAP masks without the need for a respirator–a fact that simplifies treatment. This is particularly interesting in the pre-hospital setting or in Departments with limited resources. The PEEP values used in the treatment of ACPE vary from 5 to 12.5cmH2O, depending on the study.

In applying NIPPV, a respirator is essential. For the treatment of ACPE, the PEEP values (in the same way as in CPAP) should not be less than 5cmH2O, and the inspiratory support can vary between 6 and 15cmH2O in order to secure tidal volumes of about 8ml/kg. Fig. 2 proposes an interventional ventilation algorithm for patients with ACPE.

As we have explained in this review, on choosing a ventilation mode in ACPE it has not been possible to demonstrate the superiority of one technique over the other in terms of the need for intubation, hospital mortality, hospital stay or the incidence of myocardial infarction. The technique chosen should firstly depend on equipment availability. In the equal availability of resources, and in the absence of confirmation of differences in patient outcome between the two techniques, we preferentially should use NIPPV instead of CPAP, since shortening of the time to clinical resolution in itself should be a primary objective in the management of these patients.