September - December 1999: 
Volume 12, Issue 3

Click on the image to download the Issue in PDF format.


Physiological basis of mechanical ventilation
Full text


The principal aim of mechanical ventilation is i) to correct abnormalities in arterial blood gas tension, ii) to maintain alveolar ventilation whenever the ventilatory pump (composed of the respiratory centres, motoneurons, respiratory muscles, and nerve afferents to the respiratory centres) fails, and iii) to assist the respiratory muscles in maintaining alveolar ventilation when the neuromuscular drive is normal but the mechanical load is in excess relative to the respiratory muscle force generating capacity. Mechanical ventilators can accomplish these tasks by adjusting the minute volume to correct hypercapnia and by treating hypoxemia with O2 supplement. Nevertheless, volume, frequency and timing of gas delivered to the lungs have important, disease-specific effects on cardiovascular and respiratory system function.

New technology allowed to program mechanical ventilators to deliver gas with virtually any pressure or flow profile. Significant advances have been made in producing mechanical ventilators more responsive to changes in patients ventilatory demands, and monitoring of airway pressure (Paw), volume, and flow waveforms has been developed to ameliorate ventilator functioning. Apart from the choice of inspired gas composition, it is possible to consider two other general aspects of mechanical ventilation: i) the definition of the machines mechanical output (Ventilatory mode) in order to either replace or assist the respiratory muscles, and ii) the means to ensure the machines sensing of the patient’s demand. Moreover, interactions between the patient’s ventilatory pump and the ventilator's work have to be taken into account to assess the outcome of mechanical ventilation. Finally, different syndromes leading to respiratory failure should require different ventilatory approaches to take maximum advantage of mechanical ventilation. This chapter will briefly focus on all these aspects.


The mode of mechanical ventilation is characterized by the shape of the inspiratory pressure or flow profile and determines whether a patient can increase tidal volume (VT) or rate through his or her own efforts.

Volume-preset Mode

In the volume-preset mode, each machine breath is delivered with the same inspiratory flow-time profile chosen a priori by the physician. In this mode, VT is constant and cannot be modified by the patient’s effort, since the area under a flow-time curve defines it. Volume-preset ventilation with constant (square wave) inspiratory flow is the most used. Different flow-time profiles, such as sinusoidal inspiratory flow waveforms, are sometimes used as an alternative to square wave flow in the hope of reducing the risk of barotrauma. This belief is based on the fact that peak pressure in the airways is lower in breaths with flows that decrease with increasing lung volume than with constant flows.

Four settings define the mechanical output of a ventilator operating in the volume-preset mode: i) the shape of the inspiratory flow profile, ii) VT, iii) machine rate, and iv) a timing variable in the form of either the inspiration to expiration ratio (I:E ratio), the duty cycle (TI/TTOT) where TI is the inspiratory time and TTOT is the total cycle duration, or the TI.

Pressure-preset Modes

During pressure-preset ventilation, the ventilator generates in the airways a predefined pressure (Paw) during inspiration. The resulting VT and inspiratory flow profile varies according to the mechanical properties of the respiratory system and to the strength of the patient’s inspiratory efforts. Therefore, the stiffer the lungs or chest wall, or the larger the airway resistance, or the smaller the patient’s own inspiratory efforts, the smaller will be VT. As a consequence, an increase in respiratory system workload can lead to a significant fall in minute ventilation (VE), hypoxemia, and CO2 retention, but, in contrast to volume-preset modes, it does not predispose the patient to an increased risk of barotrauma. This latter feature explains why pressure-preset ventilation is gaining popularity in the treatment of patients with acute lung injury1.

Pressure support ventilation (PSV) and pressure-controlled ventilation (PCV) are the most widely used forms of pressure-preset ventilation. During PCV, the physician sets the machine rate, the inspiratory time, and thus the I:E ratio. In PSV, switching from inspiration to expiration is linked to inspiratory flow, which in turn depends on the impedance of the respiratory system as well as the timing and magnitude of inspiratory muscle pressure output2.

PSV has become a popular weaning mode for adults. Its popularity is based on the belief that weaning from mechanical ventilation should be a gradual process and that the work of unassisted breathing through an endotracheal tube is unreasonably high and could lead to respiratory muscle failure3. Compared with weaning using synchronized intermittent mandatory ventilation (SIMV), during which volume-preset machine breaths add to spontaneous breaths, PSV is thought to offer the patient greater autonomy over inspiratory flow, VT, and inspiratory time4. In the PSV mode, a target pressure is applied to the endotracheal tube, which augments the innation pressure exerted by the inspiratory muscles (Pmus) on the respiratory system. As the lungs inflate, inspiratory flow begins to decline because Paw and Pmus are opposed by rising elastic recoil forces. When inspiratory flow reaches a threshold value, the machine turns to expiration. In most ventilators, the pressure-time profile during PSV has a square wave shape. It should be noted that those machines that take longer to reach the preset pressure plateau provide less inspiratory support.

Synchronized Intermittent Mandatory Ventilation

During SIMV, the operator (physician, respiratory therapist) is allowed to set a limited number of volume-preset breaths that are delivered every minute5. In addition, the patient is free to breathe spontaneously between machine breaths or to take breaths assisted with PSV. Volume-preset breaths are actively triggered by the patient (see below), that is, ventilator and respiratory muscle activities are "synchronised". IMV is an accepted weaning modality; even a small number of volume-preset IMV breaths per minute may make the blood gas tensions look acceptable in patients who otherwise meet criteria for respiratory failure. This evenience is likely to occur in patients with small spontaneous VT (<3 ml/kg body weight), in those with spontaneous rates of 30 breaths per minute or more, and when dyspnea and thoracoabdominal paradox indicate an increased respiratory effort6.

Mandatory Minute Ventilation

Mandatory minute ventilcition (MMV) represent a backup ventilation for patients undergoing weaning trials in the PSV mode7. The physician determines the minimum amount of VE that is to be maintained. When VE falls below this target, some ventilators increase pressure support per breath, while others switch automatically to volume-controlled assisted ventilation. Although MMV is an appealing feature of newer ventilators, its use has not been tested enough. As an example, no indication exists about the minimum VE to deliver. Similarly to IMV with low backup rates, an inappropriately low MMV setting may also determine acceptable values of arterial blood gases, but, at the same time, mask respiratory pump failure.

SIMV and MMV differ from assist-control mode (AIC) because in the latter all the patient’s inspiratory efforts are mechanically assisted by delivered breaths all characterised by either volume- or pressure-preset mode, depending on the operator choice.


The mechanical ventilation can be either controlled (controlled mechanicaI ventilation; CMV) or assisted. The difference between the two is that with CMV breaths are machine initiated, whereas with assisted modalities, the ventilator is triggered by the patient’s inspiratory efforts; in this case the machine inspiratory cycle should coincide with patient’s inspiratory phase. Occasionally, mechanicaI ventilation is considered as “controlled” when spontaneous respiratory muscle activity has been abolished by mechanical hyperventilation8 or by drugs (e.g., sedation and neuromuscular blockade).

Machine algorithms for detecting patient effort can be based on the airway pressure signal9. Because the inspiratory port of ventilators is closed during machine expiration, any inspiratory effort causes a decrease in Paw, provided that it starts near the respiratory system relaxation volume (Vr). When Paw reaches a preset threshold (usually set 1 to 2 cm H2O below the end-expiratory pressure setting), the machine switches from expiration to inspiration. In the presence of dynamic hyperinflation, the inspiratory muscles must generate considerably more pressure than the set airway trigger pressure before a machine breath is delivered10 (see also below).

Sometimes it is possible to find, especially in older ventilator models, delays of up 0.5 s between the onset of inspiratory muscles and machine response11. Sensing delays are common when the Paw is monitored in the machine rather than near the patient-ventilator interface. In this case, the ventilator tubing acts as a capacitor, delaying the transmission of pressure from the intrathoracic airway to the pressure transducer. Additional delays can be attributed to dynamic hyperinflation and physical constraints on the opening and closing of demand valves.

"Flow-by" is another algorithm for detecting patient effort. During ventilation in the flow-by trigger mode, a base flow of gas is being delivered to the patient during the expiratory as well as the inspiratory phase of the machine cycle9. Unless the patient makes an inspiratory effort, gas bypasses the endotracheal tube and flows through the expiratory machine port. In the absence of patient effort, expiratory flow is equal to inspiratory base flow. In the presence of an inspiratory effort, gas enters the patient’s lungs. As a consequence, inspiratory and expiratory base flow become different, fact that induce the ventilator to switch phase. Flow triggering has been shown to reduce the ventilatory workload upon the patients’ inspiratory muscles compared to traditional pressure-triggered systems. Moreover, application of flow triggering in mechanically ventilated patients with chronic obstructive pulmonary disease (COPD) requires less effort to initiate inspiration and provide a positive end-expiratory pressure level that is able to unload the respiratory muscles by reducing PEEPi12. With flow triggering higher minute ventilation is also obtained in COPD patients during the weaning phase12.


During CMV, in the absence of any effective muscle action, there is essentially no interaction between patient and ventilator. On the other hand, during assisted ventilation the respiratory system is simultaneously under the influence of two pumps, the patient’s own (respiratory muscles) and the ventilator; so the output of either pump can influence the output of the other and there is a considerable interaction between patient and ventilator. In the volume-preset mode, any inspiratory muscle activity cannot change either the VT or the timing of the assisted breath, the patient’s effort being spent to reduce the ventilator effort. By contrast, in the pressure-preset mode, the inspiratory muscle activity increases the alveolar to airway opening pressure gradient, allowing increases in VT. The amount of this increase will depend on respiratory mechanics of the patient, and on the resistive properties of the endotracheal tube and the respiratory circuit. The different interplay between patient’s inspiratory effort and ventilator output during volume- and pressure-preset modes can explain why non sedated patients usually prefer the latter.

Patient-ventilator interaction is described by the analysis of: i) the basic relationship between the patient-generated pressure coupled with the machine-generated pressure and the inspiratory workload (basic mechanical relation), ii) the effect of patient effort (Pmus) on ventilator pressure output (Paw), and iii) the effect of flow, volume and Paw on patient effort.

The basic mechanical relation is given by the equation of motion13:

Pmus + Paw = V x R + V x E + PEEPi
where V x R = Pres, V x E = Pel, and PEEPi = intrinsic positive end-expiratory pressure. At any instant, the forces applied to the respiratory system are the sum of Pmus and Paw that equal the opposing forces related to the resistive and elastic properties of the respiratory system. The pressure dissipated against resistance (Pres) is a function of instantaneous rate of flow. Although this resistive function is often considered as a constant and identified by a single value (resistance), representing the pressure required to generate a unit of flow. The pressure dissipated against the elastic properties at any instant (Pel) is a function of how respiratory volume is far from the relaxation volume of the respiratory system, and not only of the volume inhaled. During spontaneous breathing (Paw = 0), the progressive increase in elastic recoil during inspiration is the result of the inspiratory action of muscles, at the end of inspiratory effort, expiratory flow must begin. With volume-cycled assist, the ventilator adjusts its pressure output, such that total applied pressure remains in excess of elastic recoil until the target volume is reached; so there is no inherent coupling between the end of a patient’s effort and the start of expiratory flow. During pressure-assisted inspiration, the end of the inspiratory effort is not always synonymous of the start of expiratory flow; the predetermined Paw may be greater than the elastic recoil reached during the period of neural inspiration.

Second, the patient’s inspiratory muscle effort acts on ventilator pressure output both before and after triggering. Before triggering, in the assist mode, all ventilators require that the patient reduce Paw below the positive end-expiratory pressure (PEEP) level before assist begins. Usually the ventilator does not provide any flow until Paw decreases below a certain level (pressure triggering). In other cases, the ventilator allows air to flow in response to the decrease in Paw. Triggering occurs when flow from machine to patient exceeds a set level (flow triggering). In all case, a finite level of pressure or flow must be set to prevent false triggering. Dynamic hyperinflation has significant effects on patient-ventilator interaction before triggering. In fact:

• In the absence of dynamic hyperinflation the elastic recoil of the respiratory system is, by definition, equal to zero at the beginning of inspiration. With pressure-triggered devices, flow does not begin until Paw decreases below the pressure-trigger leveI. In this case, Paw before triggering is a direct reflection of a patient’s effort, regardless of the patient’s mechanical properties, endotracheal tube size or ventilator characteristics. By contrast, with flow-triggered devices, any inspiratory effort beginning at Vr, at which elastic recoil is zero, generates a reduction in alveolar pressure setting up a gradient of flow. The point at which Paw is measured is situated at an intermediate position between the alveoli and the ventilator. For a given pressure gradient between alveoli and ventilator, the flow generated will depend on the total resistance, the resistance of the patient’s airways of the endotracheal tube and the resistance of the ventilators adequacy as a flow-demand system. The flow generated before triggering is not a simple reflection of the intensity of a patient’s effort, but is greatly affected by factors that vary from patient to patient (airway resistance, size of endotracheal tube) and the properties of the ventilator. It is clear that a given flow threshold can be reached with different effort.

• In the presence of dynamic hyperinflation the patient must first generate enough pressure to offset the elastic recoil associated with dynamic hyperinflation (DH) before Paw can decrease or flow is generated. The DH may be stable or variable. Stable DH is when the ventilator inflation cycle does not extend much beyond the patient’s inspiratory effort. Here much of the patient’s expiratory phase is available for expiration and, even though lung volume may not reach passive FRC before the next inspiration, every inspiratory effort triggers a machine cycle and begins at the same lung volume. Variable DH develops when, following a triggered cycle, the next inspiratory effort occurs at the time when elastic recoil is still too high to allow triggering. The lack of triggering because of substantial DH in one breath allows more complete emptying in preparation for the next breath, two or three unsuccessful efforts take place before triggering occurs. This results in a variable relation between patient’s inspiratory efforts and machine cycles.

After triggering, with volume-cycled methods of support, the ventilator delivers a predetermined pattern of V and V. The required pressure is provided partly by the patient and partly by the ventilator. Machine-generated pressure bears an inverse relation to patient-generated pressure. The more the patient pulls the less the machine pushes. In a perfect volume-cycled ventilator, the relation has a slope of -1: for every 1 cm H2O of pressure generated by the patient, the machine decreases its pressure output by 1 cm H2O. With PSV, Paw is a predetermined function of time; changing effort, the patient can alter total applied pressure and influence the time course of flow and volume. With proportional assist ventilation (PAV), the relation between patient effort and Paw is positive; the more the patient pulls the more the machine pushes, an increase in patient effort causes an increase in total applied pressure and hence in flow and volume.

Third, with assisted methods of support, changes in the degree of assist will alter minute ventilation, flow, tidal volume, or airway pressure; these may have important consequences on the breathing pattern and intensity of patient effort. These kinds of interactions are mediated by:

Mechanical feedback. A greater inspiratory flow or VT will result in a smaller Pmus than would otherwise occur at the same level of muscle activation. This is due to the intrinssic properties of respiratory muscles (force-velocity and force-length relationships) as well as to geometric factors. However, the importance of this interaction seems very small, in that the flow rates and tidal volumes generated during mechanical ventilation are small relative to the physiological range of these variables.

Chemical feedback. Changes in rate or intensity of patient effort or level of assist may alter the level of alveolar ventilation and hence blood gas tension. Chemoreceptors activity is thus modulated by changes in PO2 and PCO2, fact that may affect the rate and depth of respiratory efforts. Peripheral chemoreceptors, mainly located at the bifurcation of the common carotid arteries, sense arterial O2. The CO2 tension is sensed both by the peripheral chemoreceptors and by central receptors located in the brain. Changes in PO2 and PCO2 affect both respiratory rate and intensity of effort within each breath. The effect of chemical feedback is such that ventilation would increase when blood gas tensions deteriorate and vice versa, the magnitude of this interaction depending on the ventilator mode. As a matter of fact, chemical feedback increases its effects progressively from totally controlled ventilation (CMV) to assist with volume-cycled ventilation (AMV) and, finally, to pressure-assisted ventilatory support (PSV and PAV) (i.e. proportionally to the degree of freedom allowed to the patient’s respiratory controller during mechanical ventilation). With CMV, an increase in PaCO2 as a result of an increase in metabolic rate cannot elicit any ventilatory response; the ventilator neither increases its rate nor its tidal volume in response to increases in patient’s respiratory rate or intensity of inspiratory effort. As a result PaCO2 would rise until the amount of CO2 removed by the lung matches the new level of CO2 production at the same level of ventilation. With A/C mode, the ventilator responds to only the rate component of the chemical response, without increase in tidal volume. In theory, with pressure-assisted ventilatory support the ventilator is not only responsive to patient rate, but the tidal volume obtained also varies with intensity of patient effort.

Respiratory stimuli increase ventilation usually through a combination of an increase in rate and in tidal volume. How the response is partitioned is of considerable relevance to the interaction between patient and ventilator with synchronised methods of ventilatory support. Thus, assume that a respiratory stimulant exerts its influence strictly through increasing intensity of inspiratory effort with no rate effects. In the spontaneously breathing subject, the response will be manifested as an increase in VT. In a patient on AMV, there would be no increase in ventilation, and distress may result, since the increase in VT demand is not met. With pressure-assisted methods, the increased inspiratory effort will elicit an increase in VT, thereby satisfying, partly or completely, the increase in VT demand. When the response is primarily exerted on rate, all methods of synchronised support will be able to respond.

Reflex feedback. Respiratory rate and the intensity of respiratory muscle activation are influenced by a lot of reflexes originating in the respiratory tract, and in the chest wall14-16. Since these reflexes are primarily responsive to volume and flow, they can profoundly influence the pattern and level of ventilation during pressure-assist modalities (PSV and PAV), in which the rate of breathing and tidal volume are responsive to frequency and intensity of patient effort. Tidal, as well as static changes in lung volume have important and complex effects on rate and depth of breathing efforts; these responses are mediated by vagal and chest wall receptors. Changes in tidal volume elicit reciprocal changes in the duration of inspiratory activity (Ti). As tidal volume is decreased, through an increase in mechanical load17 or in ventilator gain18, inspiratory duration increases and inspiratory activity progresses to a higher level, and vice versa. This reflex acts to limit changes in VT due to changes in pressure-assisted ventilation. In fact, in response to increases in mechanical assistance, VT increases would be limited by feedback reflexes hampering peak pressure and TI obviously, reductions in pressure assistance would lead to opposite results.

When lung emptying is delayed during expiration, as by increasing expiratory resistance, expiratory duration is prolonged, and there is usually recruitment of expiratory muscles19-21. These responses promote more complete emptying in the face of high resistance, thereby reducing the magnitude of dynamic hyperinflation. Continuous elevation of lung volume by addition of PEEP or with deliberate inflations, elicits responses similar to those of delayed emptying22,23.

• Behavioral feedback. Awake humans can intentionally modify their breathing pattern virtually in whatever direction. Usually, behavioural responses are aimed to reduce discomfort related to mechanical ventilation. As an example, patients learn quickly how to limit excessive VT during PSV by stopping their inspiratory effort just after triggering occurs, and/or by activating expiratory muscles to cycle the ventilator from inspiration to expiration. This response to PSV may explain the extreme discomfort experienced by patients when they are switched from PSV to A/C volume-preset modes in which patient’s control on ventilation is reduced or absent. Changes in volume, flow, and airway pressure are very readily perceived in awake subjects24-26.


Many diseases of the cardiopulmonary system can cause respiratory failure. It may be useful to group them into those that cause lung failure and those that cause ventilatory pump failure. The characteristic of lung failure is hypoxemia, which is usually the result of severe ventilation-perfusion mismatch. The characteristic of ventilatory pump failure is hypercapnia. Ventilatory pump failure may complicate disorders of the central nervous system, of the peripheral nerves and of the respiratory muscles. It may also accompany diseases of the lungs, such as emphysema, once the ventilatory pump fails to compensate for inefficiencies in pulmonary CO2 elimination. Two classic examples of hypoxic and hypercapnic ventilatory failure that require fundamentally different approaches to mechanical ventilation are the adult respiratory distress syndrome (ARDS) and chronic airflow obstruction. The therapeutic aim in ARDS is to increase lung volume in an attempt to reduce shunt by re-expanding collapsed and flooded alveolar units. In contrast, the therapeutic aim in a patients with hypercapnic ventilatory failure from exacerbation of airways obstruction is to reduce dynamic hyperinflation and to protect the respiratory muscles from overuse.


Acute lung injury often complicates systemic illnesses such as sepsis27, and is characterised by a dramatic impairment of pulmonary gas exchange. The general management aim in these disorders is to increase systemic oxygen delivery to cope with metabolic demands. Cardiovascular and ventilatory support are usually required to achieve this aim28.

Ventilatory support is often difficult because exceedingly high ventilatory requirements covered by mechanical ventilation expose patients at risk for barotrauma and cardiovascular collapse. Moreover, extremely rapid and shallow breathing is usually adopted by these patients, such that ventilator settings rarely can be matched, configuring a situation of severe patient-ventilator uncoupling. All of these conditions often require heavy sedation and neuromuscular blockade. Enriched O2 inspiratory mixtures, manipulation of end-expiratory lung volume, and limitation of VT are the cornerstones of the ventilatory approach of ARDS patients, even if both can have important side effects, in the former represented by oxygen toxicity29, and in the latter by barotrauma30-32.

Fractional inspired oxygen concentration

FIO2 as high as 1.0 in the first few hours of mechanical ventilation are often required to maintain adequate oxygenation, exposing patients to potential oxygen toxicity. However, the latter remains poorly defined, with exception of patients who have received bleomycin or amiodarone, drugs known to increase lungs susceptibility to oxygen radical-mediated injury33.

Manipulating end-expiratory lung volume

Non uniform lung injury is in large part responsible for ventilation-perfusion (V/Q) mismatch and shunt. The ventilatory management is therefore be directed toward re-establishing ventilation of collapsed and flooded lung regions34,35. This is obtained by increasing the distending pressure (transpulmonary pressure, PL) of these regions. There are two ways of achieving this result: raise lung volume by application of extrinsic positive end-expiratory pressure (PEEP) or increase lung volume dynamically. Since it is not uncommon for patients with acute lung injury to have a respiratory rate greater than 30 per minute, a component of dynamic hyperinflation is often present in sedated and mechanically ventilated ARDS patients36. Sedation and neuromuscular blockade usually delivered to patients add to PEEP therapy, as they abolish expiratory muscle activity. Titration of PEEP can be performed by analysis of the shape and the hysteresis area of PL-volume loops37-40. PEEP therapy is instituted to prevent derecruitment of alveoli during lung deflation and thereby reduce pulmonary shunt and V/Q mismatch35,41-43. The PEEP prevents the collapse (de-recruitment) of alveoli during lung deflation, reducing shear stresses and lung damage. It is unclear whether PEEP or dynamic hyperinflation is more effective in recruiting and preventing the collapse of edematous lung regions.

Setting tidal volume

In patients with lung injury, whose inspiratory capacity and total lung capacity (TLC) are substantially reduced and whose end-expiratory volume has been raised with PEEP, a VT of at least 10 ml/kg can have adverse effects on lung structure and function30-32,42. Barotrauma results from overdistension of lung units, not from excess pressure within the airways. Assuming that inflating the lungs to volumes above TLC is unsafe, it has become common practice to reduce VT to no more than 7 ml/kg in the management of ARDS40.

In patients with the most severe impairment in gas exchange, it is often useful to set VT as a dependent variable, resulting from end-expired and end-inspired lung volume settings. As already pointed out, adjustments in end-expired lung volume are guided by concerns for oxygenation, while tidal volume and rate are set to influence alveolar ventilation and body CO2 stores. Following this approach, the machine VT is determined by the difference between the largest volume to which the lung can safely be inflated and the smallest volume that, at the same time, keeps alveolar units from collapsing and results in an acceptable PaO2. Up to now, the definition of a safe end-inspiratory lung volume is arbitrary. Based on animal studies30,31, it seems reasonable not to exceed TLC, which in normal lungs corresponds to an alveolar pressure between 30 and 35 cm H2O and a PL of 25 cm H2O44. Peak alveolar pressure can be approximated from Pao after occlusion of the endotracheal tube at end inflation (hold or plateau pressure).

Respiratory rate

Respiratory rate (RR) setting follows VT and end-expiratory volume adjustments. Factors to be taken into account are: i) the patient’s respiratory rate, ii) the patient’s ventilatory requirement, and iii) the impact of the rate setting on breath timing. Rapid shallow breathing is a common feature of patients with hypoxic respiratory failure, such that they require rate settings over 20 breaths per minute. In the awake, non sedated patients, ventilator rates lower than this would be poorly tolerated because neurohumoral feedback from lung edema and inflammation induces rapid shallow breathing independent of chemoreceptive and mechanoreceptive effects on respiratory centres. Furthermore, in the presence of a severe gas exchange impairment, low rates and minute volumes would cause CO2 retention, adding its own effects on respiratory rate and drive. Finally, in A/C modes of ventilation, discrepancies between patient and ventilator respiratory rates would produce either extreme discomfort in control mode, or breathing patterns with inverse inspiratory to expiratory timming ratios in assisted modes. For these reasons, the machine rate should always be set close to the patient’s actual respiratory rate. If the actual rate is so high that effective ventilation cannot be achieved, then the patient needs additional sedation and possibly neuromuscular blockade.

I:E ratio

A long TI, a high TI/TTOT and a low mean inspiratory flow all promote ventilation with an inverse I:E ratio. The beneficial effects of increasing I:E beyond 1:1 on pulmonary gas exchange have proven marginal in ARDS, provided VT and end-expiratory volumes are held constant45. All ventilators provide the option of maintaining lung volume at end-inspiration for a predefined time. This time, also referred to as the end-inflation hold time or inspiratory pause time, is usually expressed as a percentage of the total cycle time (%TTOT). Long inspiratory time favours the recruitment of previously collapsed or flooded alveoli. Although alveolar recruitment could represent a desired therapeutic end point in the treatment of patients with edematous lungs, it has to be considered that the use of high volumes (and pressures) for some time may damage relatively normal units and may have adverse hemodynamic effects.

Inspiratory flow

Most ventilators require that mean inspiratory flow and its profile be specified. Increasing flow will always raise peak Pao, but this does not represent a big problem if most of the added pressure is dissipated across the endotracheal tube. Even if inspiratory flow is one of the factors that determine the regional distribution of inspired gas46, effects of flow on pulmonary gas exchange not related to changes in lung volume are somehow unpredictable, thus general guidelines are of little utility. Much more important are the effects induced by combined changes of flow, volume, and time settings on functional residual capacity and the degree of dynamic hyperinflation34,47,48.

Minute ventilation

Minute ventilation is strictly related to PaCO2 levels. In patients with acute lung injury, normocapnia can very often be achieved only with high lung inflation volumes and pressures, because they are usually hypermetabolic (high V.CO2) and in addition suffer from severe V/Q mismatch (high VD/VT). For these reasons it is not unusual to find patients with acute lung injury whose minute volume requirements exceed 20 liters per minute. The necessity to maintain an acceptable acid-base status opposes to the necessity to avoid the profound consequences of overdistension injury on lung function and on the morbidity and mortality from ARDS49. Having in mind the issue of therapeutic priorities, recent data suggest that the prevention of barotrauma precedes the goal to normalise CO2 tensions and acid-base status. The corresponding ventilation strategy has been termed "permissive hypercapnia"50. Permissive hypercapnia means that the physician accepts a PaCO2 outside the expected or "normal" range in order to minimise the potential for mechanical ventilation-induced barotrauma. Because such a ventilation strategy opposes against the limits set by the patient’s chemical feedback (see above), permissive hypercapnia usually requires heavy sedation and paralysis of the patient.


In patients with obstructive pulmonary diseases the capacity for generating expiratory flow is reduced. When obstruction is severe enough to cause ventilatory failure, dynamic airway collapse is virtually always present during the expiratory phase of the ventilator cycle10,51. These patients very often develop dynamic hyperinflation, which can have adversely effects on circulation52, may increase the risk of barotrauma48,53,54 and can place the diaphragm and inspiratory muscles at a mechanical disadvantage55. The primary therapeutic aim of mechanical ventilation in obstructive lung disease is to minimise the thoracic volume about which the lungs are ventilated. Patients with chronic obstruction from emphysema or chronic bronchitis (unless they are fighting the ventilator) are usually easy to ventilate and may simply need respiratory muscle rest and a resetting of CO2 response thresholds to more normal values. These secondary therapeutic objectives are highly controversial. In contrast, patients with acute asthma (or status asthmaticus) often fight the ventilator and therefore often require sedation, neuromuscular blockade, and ventilation with permissive hypercapnia50,54. Steroids and paralytic agents in such patients may cause muscular damage, and may require prolonged mechanical ventilation for weakness long after lung mechanics normalise56.

Minimizing dynamic hyperinflation

Dynamic hyperinflation is associated with an increase in alveolar pressure at end expiration. This pressure, also called intrinsic positive end expiratory pressure (PEEPi), auto-PEEP, or inadvertent PEEP, reflects the elastic recoil of the respiratory system at end expiration plus any pressure generated by respiratory muscles10,52. In the absence of muscle activity, the degree of dynamic hyperinflation can be inferred from the endexpiratory airway occlusion pressure (PEEPi), and the elastance of the relaxed respiratory system (Ers):

Vee - Vr = Ers/PEEPi
where Vee = volume of lungs at end expiration. Alternatively, it can be also measured directly by allowing the patient freely exhale through a pneumotachometer connected to an open expiratory line till Vr is reached. This latter maneuver can require till to 30-45 seconds.

Ventilator adjustments designed to minimise dynamic hyperinflation should be aimed to lower mean expiratory flow (VT/TE). In a paralysed asthmatic patient, VT can be reduced to as little as about 5 ml/kg, while TE is prolonged through increases in mean inspiratory flow (1 to 1.5 L/s), adjustments in the I:E (1:4 or 1:5), and reductions in machine backup rate (=10 breaths per minute). This strategy results in a significant reduction of VE and alveolar ventilation, thus producing hypercapnia. However, even acidemia to pH values of about 7.20 is usually well tolerated in paralyzed subjects57. High inspiratory flow settings, which are required to prolong TE, can increase peak Paw and may raise concerns about barotrauma. It must be emphasised, however, that much of this added "resistive pressure" is dissipated along the endotracheal tube and proximal airways and that on balance, increasing the rate of lung inflation seems less damaging than ventilating the lungs near TLC54.

Permissive hypercapnia and neuromuscular blockade are rarely required in patients with ventilatory failure from exacerbations of chronic obstructive lung diseases. Nevertheless, many patients have high respiratory rates, making it difficult to prolong TE in order to allow them to breathe near Vr51. Because in the non paralysed subject hypercapnia sets limits to the reduction in VT, attempts must be made to reduce the patient’s respiratory rate. Sometimes the only way to minimise hyperinflation avoiding the use of paralysing agents is the judicious use of sedatives to decrease inspiratory drive to the point at which inspiratory efforts fail to initiate a machine breath.

Use of continuous positive airway pressure

The use of positive end-expiratory pressure (PEEP) during mechanical ventilation and of continuous positive airway pressure (CPAP) during spontaneous breathing (e.g. weaning) for the treatment of acute respiratory failure (ARF) represents a widely accepted procedure in the Intensive Care Unit (ICU). Benefits of this treatment relate to the improved gas exchange and lung mechanics consequent to the recruitment of previously unventilated, perfused, areas (see above). On the other hand, CPAP/PEEP may be responsible for some detrimentaI effects on respiratory mechanics such as alveolar overdistension, decreased compliance and eventually pneumothorax (SB), as well as on hemodynamics, since it may reduce venous return and impair the cardiac function58-61.

Because the effects of CPAP and PEEP depend on lung distension, their use has been traditionally adversed for the treatment of acute respiratory failure in patients with chronic obstructive pulmonary disease (COPD). The rationale of this choice lies on the fact that pulmonary hyperinflation is common in those patients. In this condition, it has been speculated that PEEP/CPAP might: i) worsen the impaired function of the inspiratory muscles by further reducing their operational length because of the increase in lung and chest volume, ii) be responsible of barotrauma, and iii) determine hemodynamic compromise. Furthermore, advantages would be modest, as hypoxia is usually reversed with low concentrations of oxygen in patients with COPD, even during acute exacerbation. However, in contrast to the above consolidated point of view, there is increasing evidence that PEEP during assisted ventilation62, and CPAP during spontaneous breathing63 are of benefit in patients with acute exacerbation of COPD. During controlled mechanical ventilation, conflicting results have been obtained. Tuxen64 reported, in paralyzed patients with asthma, an increase in lung volume with PEEP, associated with hypotension and evidence of reduced oxygen delivery. At variance, Ranieri and colleagues65 found that the adverse effects of PEEP on hemodynamics and lung volume were detectable only with PEEP exceeding 80% of the initial PEEPi. Similar results were obtained also by Rossi et al.66 Dynamic pulmonary hyperinflation due to expiratory flow limitation, a common feature in this kind of patients, can explain the apparent contrast between older and newer positions. In normal subjects, lung volume at the end of a tidal expiration (EELV) represents the relaxation volume (Vr) of the respiratory system67. In contrast, in patients with airflow limitation, the end-expiratory lung volume may be well above Vr, because the rate of lung emptying is decreased, and expiration is curtailed by the start of the next inspiration before full decompression of the lung to the static relaxation volume of the respiratory system. Under these circumstances, i.e. dynamic pulmonary hyperinflation and passive expiration, some elastic recoil is still present at end-expiration and positive end-expiratory pressure (PEEPi) can be recorded by closing airways at the end of a relaxed exhalation.

PEEPi strongly challenges inspiratory muscles since it acts as an "inspiratory threshold load" which has to be offset before a negative pressure in the central airway and hence tidal volume can be generated68. This extra-burden imposed on inspiratory muscles can be very high. When PEEPi is due to flow limitation69, expiratory flow does not depend on the gradient between alveolar pressure and atmosphere, but it is determined by the critical transmural pressure at the point in which airways collapse. In this condition the application of an external positive pressure at the airway opening (CPAP/PEEP) does not affect alveolar pressure, and hence lung volume, until a critical value, somewhat lower than intrinsic PEEP is exceeded63,69,70. This particular feature allows CPAP to work during inspiration, in which the addition of a positive pressure downstream the site of critical closure counterbalances the threshold load (i.e. PEEPi), thus reducing the inspiratory effort, without adverse effects linked to overinflation during expiration that express themselves only if the critical pressure is approached69,70. These mechanisms have been validated in intubated patients with COPD during weaning63,71. CPAP was demonstrated to be effective in reducing inspiratory effort without increasing the end-expiratory lung volume untill a criticaI value of positive pressure was exceeded. More interestingly, CPAP set at 80%-90% the level of PEEPi during spontaneous breathing was demonstrated to significantly reduce the inspiratory effort without increasing EELV in non-intubated patients with acute exacerbations of COPD72, and in ventilator-dependent, tracheostomized COPD patients73.

However, some constraints suggest not to apply CPAP indiscriminately. First, intrinsic PEEP can be present also in the absence of flow limitation36,69,74. For example, if severe airflow obstruction is associated to high respiratory rate, the expiratory time can be too short to allow the respiratory system to reach its relaxation volume. In this condition, expiratory flow is driven throughout expiration by the difference in pressure between the alveoli and the airway opening52. It follows that the expiratory flow (in absence of actively contracting expiratory muscles) decreases, and end-expiratory lung volume increases when any level of positive pressure is applied69, thus enlarging the risk of volume-related adverse elfects. Display of the volume-flow (VF) relationship during a relaxed expiration provides a simple, though qualitative, tool to assess flow limitation75, thus helping physicians to decide whether CPAP can be safely and usefully applied.

Second, the level of intrinsic PEEP can vary greatly among flow limited patients63,69,75-77. With this respect, standard levels of CPAP, for example 5-10 cmH2O, seem inappropriate, since they could be well above or below actual PEEPi. High levels of CPAP are not warranted; on the other hand, excessively low levels of CPAP reduce PEEPi only marginally and, hence, are of limited or no benefit at all. Therefore, the measurement of PEEPi on an individual basis helps to define the level of CPAP or PEEP (as above mentioned sligthly below PEEPi) required to optimize the benefits and to minimise the adverse effects65,72. The technique employed to measure PEEPi during spontaneous breathing and assisted ventilation in patients, who are often dyspneic and badly tolerate manipulations, may be considered by someone quite invasive, since it requires the positioning of esophageal and gastric balloon-tipped catheters72,76. Alternatively, inductive plethysmography has been recently introduced to monitor the changes in EELV consequent to the application of positive airway pressure in dynamically hyperinflated patients78,79. This approach, being non-invasive, has the advantage that can be repeated very frequently, allowing a fine adjustment of therapy in unstable patients.

Third, it must be clearly stated that CPAP, as well as PEEP, are supporting treatments, and that they are ineffective on the accompanying hyperinflation, which, per se, has detrimental effects on the inspiratory muscles. At high lung volume inspiratory muscle function is greatly impaired, since operational length of the inspiratory muscles is shifted to the left of the force-length relationship (i.e. muscle fibers are shorter than the "optimal" length) determining less force per unit of contraction. Moreover, geometric factors (decreased appositional area) reduce the effectiveness of shortening80. Thus, other therapeutic measures (i.e. bronchodilator and anti-inflammatory drugs, drainage of secretions etc.) must be associated to reduce PEEPi, as well as dynamic hyperinflation.


1. Marry TW, Marini JJ. Inverse ratio ventilation in ARDS: rationale and implementation. Chest 1991, 100:494-504.
2. Maclntyre NR. Ho L-I: Effects of initial flow rate and breath termination criteria on pressure support ventilation. Chest 1991, 99:134-8.
3. Brochard L, Pluskwa F, Lemaire F. Improved efficacy of spontaneous breathing with inspiratory pressure support. Am Rev Respir Dis 1987, 136:411-5.
4. Maclntyre NR. Respiratory function during pressure support ventilation. Chest 1986, 89:677-83.
5. Weisman lM, Rinaldo JE, Rogers RM, Sanders MH. Intermittent mandatory ventilation. Am Rev Respir Dis 1983, 127:641-7.
6. Weiss JW, Rossing TH, Ingram LTH Jr. Effect of intermittent mandatory ventilation on respiratory drive and timing. Am Rev Respir Dis 1983, 127:705-8.
7. Sassoon CSH. Positive pressure ventilation: alternate modes. Chest 1991, 100:1421-9.
8. Prechter GC, Nelson SB, Hubmayr RD. The ventilatory recruitment threshold for carbon dioxide. Am Rev Respir Dis 1990, 141:758-64.
9. Sassoon CSH. Mechanical ventilation design and function: the trigger variable. Respir Care 1992, 37:1056-69.
10. Rossi A, Gottfried SB, Zocchi L, Higgs BD, Lennox S, Calverley PMA, Grassino A, Milic-Emili J. Measurernent of static compliance of the total respiratory system in patients with acute respiratory failure during mechanical ventilation: the effect of "intrinsic" PEEP. Am Rev Respir Dis 1985, 131:672-7.
11. Gurevitch MJ, Gelmont D. Importance of trigger sensitivity to ventilator response delay in advanced chronic obsttuctive pulmonary disease with respiratory failure. Crit Care Med 1989, 17: 354-9.
12. Ranien VM, Mascia L, Petruzzelli V, Bruno F, Bnenza A, Giuliani R. Inspiratory effort and measurement of dynamic intrinsic PEEP in COPD patients: effects of ventilator triggering systems. Intensive Care Med 1995, 21:896-903.
13. Mead J, Agostoni E. Dynamics of breathing. In: Handbook of Physiology, Section 3. Respiration. Vol. 1. Edited by WO Fenn and H Rahn. Washington DC, American Physiological Society, 1964:411-27.
14. Coleridge HM, Coleridge JCG. Reflexes evoked from tracheobronchial tree and lungs. In: Handbook of Physiology: The respiratory system, VoI. 2. Edited by NS Cherniack and JG Widdicombe. Bethesda, MD, American Physiological Society, 1986: 395-430.
15. Shannon R. Reflexes from respiratory muscles and costovertebral joints. In: Handbook of Physiology: The respiratory system, Vol. 2. Edited by NS Chemiack and JG Widdicombe. Bethesda, MD, American Physiological Society, 1986: 431-48.
16. Younes M. Remmers J. Control of tidal volume and respiratory frequency. In: Regulation of breathing. Edited by TF Hombein. New York, Marcel Dekker, 1981: 621-71.
17. Grunstein MM, Younes M, Milic-Emili J: Control of tidal volume and respiratory frequency in anesthetized cats. J Appl Physiol 1973, 35:463-76.
18. Claik FJ, von Euler C: On the regulation of depth and rate of breathing. J Physiol 1972, 222:267-95.
19. Zechman FW, Frazier DT, Lally DA: Respiratory volume-time relationships duling resistive loading in the cat. J Appl Physiol 1976, 40:177-83.
20. Remmers JE, Bartlett D Jr: Reflex control of expiratory airflow and duration. J Appl Physiol 1977, 42:80-7.
21. Gautier H, Remmers JE, Bartlett D Jr. Control of the duration of expiration. Respir Physiol 1973, 18:205-21.
22. Bishop B, Bachofen H. Vagal control of ventilation and respiratory muscles during elevated pressures in the cat. J Appl Physiol 1972, 32:103-12.
23. Grunstein M, Wyszobrodski l, Milic-Emili J: Regulation of frequency and depth of breathing during expiratory threshold loading in cats. J Appl Physiol 1975, 38:869-74.
24. KIllian KJ, Campbell EJM. Dyspnea. In: The thorax. Edited by C Roussos and PT Macklem. New York, Marcel Dekker, 1985:787-928.
25. Altose MD. Dyspnea. In: Current pulmonology, Vol. 7. Edited by DH Simmons. Chicago, Year Book Medical Publishers, 1986:199-226.
26. Zechman FW, Wiley RL. Afferent inputs to breathing: Respiratory sensation. In: Handbook of Physiology. Section 3. The respiratory system, VoI. 2. Pt 1. ControI of breathing. Bethesda, MD. American Physiological Society, 1986:449-74.
27. Bone RC, Balk ItA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RMH, Sibbald WI (ACCP/SCCM Consensus Conference Committee): Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 1992, 101:1644-55.
28. Wood LDH, Prewitt RM: Cardiovascular management in acute hypoxemic respiratory failure. Am I Cardiol 1981, 47:963-72.
29. Jackson RM. Pulmonary oxygen toxicity. Chest 1985, 88:900-5.
30. Dreyfuss D, Basset G, Soler P, Saumon G. Intermittent positive-pressure hyperventilation with high inflation pressures produces pulmonary microvascular injury in rats. Am Rev Respir Dis 1985, 132:880-4.
31. Kolobow T, Moretti MP, Fumagalli R, Mascheroni D, Prato P, Chen V, Joris M. Severe impairment in lung function induced by high peak airway pressure during mechanical ventilation. Am Rev Respir Dis 1987, 135:312-5.
32. Macklin MT, Macklin CC. Malignant interstitial emphysema of the lungs and mediastinum as an important occult complication in many respiratory diseases and other conditions: an interpretation of the clinical literature in the light of laboratory experiment. Medicine 1944, 23:281-358.
33. Rosenow EC III, Myers IL, Swensen SJ, Pisani RI, Drug-induced pulmonary disease: an update. Chest 1992, 102:239-50.
34. Hubmayr RD, Abel MD, Rehder K. Physiologic approach to mechanical ventilation. Crit Care Med 1990, 18:103-13.
35. Malo J, Ali J, Wood LDH. How does positive end-expiratory pressure reduce intrapulmonary shunt in canine pulmonary edema? J Appl Physiol 1984, 57:1002-10.
36. Brosegliini C, Brandolese R, Poggi R, Bernasconi M, Manzin E, Rossi A. Respiratory resistance and intrinsic positive end-expiratory pressure (PEEPi) in patients with the adult respiratory distress syndrome (ARDS). Eur Respir 1988, 1:726-31.
37. Falke KJ, Pontoppidan H, Kumar A, Leith DE, Geffin B, Laver MB. Ventilation with positive end-expiratory pressure in acute lung disease. Clin Invest 1972, 51:2315-23.
38. Matamis D, Lemaire F, Harf A, Brun-Buisson C, Ansquer quer JC, Atlan G. Total respiratory pressure-volume curves in the adult respiratory distress syndrorne. Chest 1984, 86:58-66.
39. Ranieri VM, Eissa NT, Corbeil C, Chasse M, Braidy I, Matar N, Milic-Emili I. Effects of positive end-expiratory pressure on alveolar recruitment and gas exchange in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1991, 144:544-51.
40. Amato MBP, Barbas CSU, Medeiros DM, et aI. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998, 338:347-54.
41. Argiras EP, Blakely CR, Dunnill MS, Ostremski S, Sykes MK. High PEEP decreases hyaline membrane formation in surfactant deficient lungs. Br J Anaesth 1987, 59:1278-85.
42. Corbridge TC, Wood LDH, Crawford GP, Chudoba MJ, Yanos l, Sznajder JI. Adverse effects of large tidaI volume and low PEEP in canine acid aspiration. Am Rev Respir Dis 1990, 142:311-5.
43. Dreyfuss D, Soler P, Basset G, Saumon G. High inflation pressure pulmonary edema: respective effects of high airway pressure, high tidal volume, and positive end-expiratory pressure. Am Rev Respir Dis 1988, 137: 1159-64.
44. Agostoni E, Hyatt RE: Static behavior of the respiratory system. In: Macklem PT, Mead l, editors. Handbook of physiology, vol 3. Baltimore: Waverly Press, 1986:113-30.
45. Cole AGH, Weller SF, Sykes MK. Inverse ratio ventilation compared with PEEP in adult respiratory failure. Intensive Care Med 1984, 10:227-32.
46. Otis AB, Mckerrow CB, Bartlett RA, Mead J, Mcllroy MB, Selverstone NJ, Radford EP Jr. MechanicaI factors in distribution of pulmonary ventilation. J Appl Physiol 1956, 8:427-43.
47. Gay PC, Rodarte JR, Tayyab M, Hubmayr RD: The evaluation of bronchodilator responsiveness in mechanically ventilated patients. Am Rev Respir Dis 1987, 136:880-5.
48. Tuxen DV, Lane S. The effects of ventilatory pattern on hyperinflation, airway pressures, and circulation in mechanical ventilation of patients with severe airflow obstruction. Am Rev Respir Dis 1987, 136:872-9.
49. Sykes MK: Does mechanical ventilation damage the lung? Acta Anaesthesiol Scand 1991, 35(suppl 95):35-9.
50. Darioli R, Perret C. Mechanical controlled hypoventilation in status asthmaticus. Am Rev Respir Dis 1984, 129:385-7.
51. Reinoso MA, Gracey DR, Hubmayr RD. Interrupter mechanics of patients admitted to a chronic ventilator dependency unit. Am Rev Respir Dis 1993, 148:127-31.
52. Pepe PE, Marini JJ: Occult positive end-expiratory pressure in mechanically ventilated patients with airflow obstruction: the auto-PEEP effect. Am Rev Respir Dis 1982, 126:166-70.
53. Tuxen DV, Williams TJ, Scheinkestel CD, Czamy D, Bowes G. Use of a measurement of pulmonary hyperinflation to control the level of mechanical ventilation in patients with acute severe asthma. Am Rev Respir Dis 1992, 146:1136-42.
54. Williams TJ, Tuxen DV, Scheinkestel CD, Czamy D, Bowes G. Risk factors for morbidity in mechanically ventilated patients with acute severe asthma. Am Rev Respir Dis 1992, 146:607-15.
55. Hubmayr RD, Sprung J, Nelson S. Determinants of transdiaphragmatic pressure in dogs. J Appl Physiol 1990, 69:2050-6.
56. Isenstein DA, Venner DS, Duggan J. Neuromuscular blockade in the intensive care unit. Chest 1992, 102:1258-66.
57. Capellier G, Toth lL, Walker P, Marshall I, Winton T, Demajo W. Hemodynamic effects of permissive hypercapnia. Am Rev Respir Dis 1992, 145:A527.
58. Levine M, Gilbert R, Auchincloss JH Jr. A comparison of the effects of sighs, large tidal volumes, and positive end expiratory pressure in assisted ventilation. Scand J Respir Dis 1972, 53:101-8.
59. Uzawa T, Asbaugh DG. Continuous positive-pressure breathing in acute hemorrhagic pulmonary edema. J Appl Physiol 1969, 26:427-32.
60. Lutch JS, Murray JF. Continuous positive-pressure ventilation: effects on systemic oxygen transport and tissue oxygenation. Ann Intern Med 1972, 76:193-202.
61. King EG, Jones RL, Patakas DA. Evaluation of positive end-expiratory pressure therapy in the adult respiratory distress syndrome. Can Anaesth Doc J 1973, 20:546-58.
62. Smith TC, Marini JJ: lmpact of PEEP on lung mechanics and work of breathing in severe airflow obstruction. J Appl Physiol 1988, 65(4):1488-99.
63. Petrof BJ, Legare M, Goldberg P, Milic-Emili J, Gottfried SB: Continuous positive airway pressure reduces work of breathing and dyspnea during weaning from mechanical ventilation in severe chronic obstructive pulmonary disease. Am Rev Respir Dis 1990, 141:281-9.
64. Tuxen DV. DetrimentaI effects of positive end-expiratory pressure during controlled mechanical ventilation of patients with severe airflow obstruction. Am Rev Respir Dis 1989, 140:5-9.
65. Ranieri VM, Giuliani R, Cinnella G, Pesce C, et al: Physiologic effects of positive end-expiratory pressure in patients with chronic obstructive pulmonary disease during acute ventilatory failure and controlled mechanical ventilation. Am Rev Respir Dis 1993, 147:5-13. 66. Rossi A, Polese G, Brandi G, Conti G: Intrinsic positive end-expiratory pressure (PEEPi). Intensive Care Med 1995, 21:522-36.
67. Vineoar A, Sinnett EE, Leith DE: Dynamic mechanisms determine functional residual capacity in mice, Mus musculus. J Appl Physiol 1979, 46:867-71.
68. Milic-Emili J, Gottfried SB, Rossi A. Dynamic hyperinflation: intrinsic PEEP and its ramifications in patients with respiratory failure. In: Intensive Care Med. JL Vincent ed, 1987:192-8.
69. Gav PC, Rodarte JR, Hubmayr RD. The effects of positive expiratory pressure on isovolume flow and dynamic hyperinflation in patients receiving mechanical ventilation. Am Rev Respir Dis 1989, 139:621-6.
70. Tobin MJ, Lodato RF. PEEP, auto-PEEP, and waterfalls. Chest 1989, 96:449-51.
71. Sassoon CSH, Light RW, Lodia R, Sieck GC, Mahutte CK. Pressure-time product during continuous positive airway pressure, pressure support ventilation, and T-piece during weaning from mechanical ventilation. Am Rev Respir Dis 1991, 143:469-75.
72. Appendini L, Patessio A, Zanaboni S, Carone M, Gukov B, Donner CF, Rossi A. Physiologic effects of positive end-expiratory pressure and mask pressure support during exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1994, 149:1069-76.
73. Appendini L, Purro A, Patessio A, Zanaboni S, Carone M, Spada E, Donner CF, Rossi A. Partitioning of inspiratory muscle workload and pressure assistance in ventilator-dependent COPD patients. Am J Respir Crit Care Med 1996, 154:1301-9.
74. Brandolese R, Broseghini C, Polese G, et aI. Effects of intrinsic PEEP on pulmonary gas exchange in mechanically-ventilated patients. Eur Respir J 1993, 6:358-63.
75. Rossi A, Brandolese R, Milic-Emili J, Gottfried SB. The role of PEEP in patients with chronic obstructive pulmonary disease during assisted ventilation. Eur Respir J 1990, 3:818-22.
76. Dal Vecchio L, Polese G, Poggi R, Rossi A. "Intrinsic" positive end-expiratory pressure in stable patients with chronic obstructive pulmonary disease. Eur Respir J 1990, 3:74-80.
77. Brochard L, Isabey D, Piquet J et aI. Reversal of acute exacerbations of chronic obstructive lung disease by inspiratory assistance with a face mask. N Engl J Med 1990, 323:1723-30.
78. Hoffman RA, Ershowsky P, Krieger BP. Determination of auto-PEEP during spontaneous and controlled ventilation by monitoring changes in end-expiratory thoracic gas volume. Chest 1989, 96:613-6.
79. Valta P, Takala J, Foster R, Weissman C, Kinney JM. Evaluation of respiratory inductive plethysmography in the measurement of breathing pattern and PEEP-induced changes in lung volume. Chest 1992, 102:234-8.
80. Derenne JPh, Fleury B, Pariente R. Acute respiratory failure of chronic obstructive pulmonary disease. Am Rev Respir Dis 1988, 138:1006-33.