Mechanical Ventilation

Chapter 8


Mechanical Ventilation




Introduction


Mechanical ventilation (MV) is an important tool for resuscitation of critically ill patients in the emergency department (ED). It is vital that ED practitioners have a thorough understanding of the basics of MV and to know when to apply these principles and how to support patients in respiratory or cardiac failure. Hospital overcrowding has led to a delay in transfer of mechanically ventilated patients out of the ED, and ventilator management often falls on the emergency medicine physician. Also, during nights and weekends in some facilities, the ED physician may be called on to troubleshoot or stabilize mechanically ventilated patients in the intensive care unit (ICU). The traditional view of MV as a prescription that fits virtually all patients equally should be discarded as a gross misunderstanding of pulmonary pathophysiology. Increasing evidence has shown that the mechanism of lung ventilation by MV may be as deleterious as it is helpful.1 Because patients remain in the ED while mechanically ventilated, ED clinicians should embrace the established paradigm of pulmonary-protective MV strategies as a cornerstone of care.



Basic Physiology


Understanding basic pulmonary physiology is essential to understanding how to initiate MV. This ensures that the method of gas delivery meshes with the patient’s underlying physiology to avoid ventilator-induced lung injury.



Minute Volume and Alveolar Ventilation


The volume of air that moves in and out of a patient’s lungs per minute is termed minute volume (image). image is the product of tidal volume (Vt) and respiratory frequency or rate (f):



image


Normal image is 7 to 10 L/min. Vt can be further broken down into alveolar volume (Va) and dead space volume (Vds):



image


In healthy young persons, the anatomic dead space is accounted for by the trachea and the larger airways and is approximately 2.2 mL/kg lean body weight. In disease states, in addition to the anatomic dead space, there is a variable amount of “pathologic” dead space, which corresponds to ventilated alveoli and respiratory bronchioles that are not adequately perfused. The sum of anatomic and pathologic dead space is often referred to as physiologic dead space.


Alveolar minute ventilation (image) is the product of rate times Vt minus dead space:



image


image and the rate of CO2 production by the body determine the partial pressure of CO2 in the alveoli (Paco2), which is approximately equal to systemic arterial CO2 tension.



Volume-Pressure Relationship


Volume and pressure are related for a given respiratory system. A given volume (V) will create a certain pressure (P) relative to the compliance (C) of the respiratory system. The respiratory system consists of the ventilator tubing, endotracheal (ET) tube, trachea, airways, lung parenchyma, chest wall, and diaphragm tension. For example, a 500-mL volume will create a certain pressure based on the compliance of the respiratory system. Increasing the volume will increase the pressure in the system. Decreasing the volume will result in lower pressure. Decreasing the compliance (i.e., making the system “stiffer”) will increase the pressure in the system. Increasing the compliance will decrease the pressure in the system.


image


Conversely, this relationship also holds for volume. For example, a pressure of 20 cm H2O will create a certain volume based on the compliance of the system. Increasing pressure will result in higher volume. Decreasing pressure will result in lower volume. Decreasing compliance will result in lower volume. Increasing compliance will result in higher volume.


image



Airway Pressures



Plateau Pressure


Plateau pressure is measured at the end of inspiration with a short breath-hold (Fig. 8-1). At this point no airflow should be occurring. This is considered a static pressure. By understanding the aforementioned volume-pressure relationship, one can easily deduce how plateau pressure is inversely related to respiratory system compliance and directly related to volume. Anything that decreases compliance will increase plateau pressure. Increasing compliance will decrease plateau pressure. Decreasing volume will decrease plateau pressure (a major tenet in lung-protective ventilation).




Peak Airway Pressure


Peak airway pressure is derived during inspiration and thus incorporates airflow. Because there is air movement during this measurement, it is considered a dynamic pressure. It reflects the dynamic compliance of the entire respiratory system and incorporates static compliance and airflow. Peak airway pressure can never be lower than plateau pressure. In addition, because its main distinction is that it incorporates airflow, it is reflective of resistance to airflow. Anything that decreases compliance or increases resistance to airflow will increase peak airway pressure. Increasing compliance or decreasing resistance to airflow will decrease peak airway pressure.


A physiologically appropriate means of detecting and monitoring bronchospasm is the peak-plateau gradient. A normal gradient is less than 4 cm H2O pressure, and elevated values indicate increased airway resistance. The efficacy of treatment with β2-agonists, steroids, intravenous magnesium, or diuresis may be assessed by monitoring the changes in this gradient (Fig. 8-2).




Positive End-Expiratory Pressure


Positive end-expiratory pressure (PEEP) is the pressure in the airway at the end of exhalation. PEEP helps keep the large noncartilaginously supported airways and the smaller alveoli open to prevent collapse, atelectasis, and ensuing hypoxia. The ventilation required to compensate for this triad commonly worsens lung compliance and is associated with ventilator-induced lung injury. Progressive increases in PEEP result in elevations in both total lung pressure and total lung volume. For example, serial elevations in PEEP often result in increased plateau pressure and elevated functional residual capacity (FRC; i.e., lung volume).



Extrinsic PEEP: When discussing MV and PEEP, most often authors are referring to extrinsic PEEP (PEEPe). This is also referred to as applied PEEP. It is the PEEP that is extrinsically applied by the ventilator. When PEEP is used without a subscript in this chapter, it refers to PEEPe. The useful PEEP range is from 3 to 20 cm H2O.2 PEEP is used to increase FRC and move the zero pressure point of each alveolar unit more proximally in the airway and thereby prevent early alveolar collapse.3 By so doing, PEEP increases the available number of alveolar units that can participate in gas exchange. The primary effect of PEEP on gas exchange is improvement in oxygenation, not removal of CO2. CO2 clearance is rather efficient and will be well preserved, even in hypoxic situations. By opening one alveolar unit, the tendency of adjacent units is to open as well (i.e., alveolar codependency) (Fig. 8-3).4 Excessive PEEP will compromise hemodynamics. There are two primary questions to ask when using PEEP to augment oxygenation: (1) What is the “optimal PEEP” and (2) Is the current amount of PEEP compromising the patient’s hemodynamics?



PEEP is not without untoward side effects, and increased levels of PEEP can lead to lung injury and hemodynamic compromise.5 Increased intrathoracic pressure can result in cardiac compression and collapse, principally of the right atrium. It is imperative that the patient be adequately volume-resuscitated because preload depletion compounds this problem. Desired levels of PEEP simply may not be possible because of deleterious effects on cardiac output.


Optimal PEEP can be determined in several ways. One is to increase PEEP until there are no longer increases in Po2. However, this method may result in several untoward events. First, oxygen tension may increase steadily, but carbon dioxide pressure (Pco2) may increase as a result of alveolar overdistention. With overdistention, alveolar pressure may exceed pulmonary arteriolar pressure and actually decrease pulmonary blood flow and clearance of CO2. Second, alveolar overdistention may increase total intrathoracic pressure and result in diminished venous return and cardiac output. Third, decreased venous return may cause cerebral venous hypertension. The optimal PEEP for one organ system may be deleterious for another. For example, the optimal PEEP for ideal oxygenation may be the worst PEEP for cerebral venous drainage.


An alternative is to increase PEEP until a complication of PEEP occurs (e.g., elevation in Pco2, hypotension) and then reduce PEEP if needed (inability to tolerate hypercapnia) or expand the patient’s intravascular volume to combat the decreased venous return.


Another excellent method of determining the optimal PEEP is guided by assessing changes in plateau pressure with changes in PEEP. As PEEP is increased from a minimal level, the patient’s peak airway pressure and plateau pressure will increase by the amount of PEEPe. When the optimal PEEP for the lung units is achieved, plateau pressure will no longer increase. As the lung is optimally recruited, peak and plateau pressure may decrease because more volume of lung is available to receive a set Vt. Once this level is exceeded, there will be further increases in plateau pressure beyond the incremental increase in PEEP as the units overdistend. Therefore, the clinician must readily identify the plateau in this plateau pressure trend. The same relationship may be displayed graphically in the dynamic pressure-volume loop (Fig. 8-4). The lower limb of the loop represents the pressure required to open the alveolar units.6,7 In the absence of PEEP (or inadequate PEEP), this limb is prolonged and flattened and has an inflection point far to the right of the origin of the loop (Fig. 8-5). As PEEP is progressively increased, the inflection point travels to the left. When the optimal PEEP is achieved, there will be a rapid upstroke of the loop because the vast majority of the functional lung units are already open and ready to be ventilated (see Fig. 8-4). This strategy is known as the open lung model of MV.6




Irrespective of what technique is used, it is currently widely agreed that plateau pressure should not exceed 30 cm H2O. If respiratory system compliance is so low that plateau pressure exceeds 30 cm H2O, either PEEP or Vt has to be decreased. If this is not possible because of either recalcitrant hypoxia or acidosis, rescue therapies may need to be used (see the section “Acute Lung Injury and Acute Respiratory Distress Syndrome”).



Intrinsic PEEP: Intrinsic PEEP (PEEPi) is additional pressure that is generated within the airways from trapped gas that should have been exhaled but for various reasons (commonly obstruction to exhalation such as in chronic obstructive pulmonary disease [COPD]) was not. PEEPi is also referred to as auto-PEEP, dynamic hyperinflation, and breath stacking. For the remainder of this chapter it will be referred to as PEEPi.


PEEPi can cause hemodynamic instability secondary to decreased venous return, just like high levels of PEEP.7 PEEPi may be detected in two ways: (1) evaluation of the flow-time trace or (2) disconnection of the patient from the ventilator and listening for additional exhaled gas after an exhalation should have occurred.6 The flow-time trace will demonstrate that the exhalation is not yet completed before the next breath has been initiated (Fig. 8-6).8





Equipment—standard Options


Perhaps one of the most confusing aspects of MV is the plethora of terms and acronyms that are used. Understanding the basic terminology helps clarify this subject. The following discussion explores machine features and settings. Regardless of which ventilator is used, a limited number of standard features are common to each (Fig. 8-7).




Set Respiratory Rate


Most ventilators allow the clinician to set a respiratory rate. The respiratory rate and actual Vt determine a patient’s minute ventilation. Patients intubated for airway protection because of trauma or toxicosis often do well with a normal minute ventilation. Initially setting the respiratory rate at 10 to 14 breaths/min and Vt at 7 to 8 mL/kg ideal body weight (IBW) is usually sufficient. Adjustments can be made based on arterial blood gas (ABG) analysis, end-tidal CO2, or venous blood gas and pulse oximetry. Patients who are septic or have severe acidosis often require higher minute ventilation. Respiratory rates can be increased, as can Vt, but volumes higher than 10 mL/kg IBW should not be used because of the risk of inducing ventilator-associated lung injury. In special scenarios such as acute lung injury (ALI) and acute respiratory distress syndrome (ARDS), initial Vt values should be lowered to 6 mL/kg IBW within 2 hours of intubation. Some of these specific scenarios are discussed later.



Fraction of Inspired Oxygen


All ventilators can deliver an adjustable fraction of inspired oxygen (Fio2). Recommendations are to set it initially at 1.0 because the act of transitioning from negative pressure ventilation (normal physiologic breathing) to positive pressure ventilation (PPV) may unpredictably alter ventilation-perfusion (image) matching. Although initially an Fio2 of 100% is optimal, it is beneficial to quickly titrate Fio2 down because of the theoretical risk for oxygen toxicity. Make adjustments based on ABG analysis or pulse oximetry, with a goal of keeping arterial Po2 higher than 60 mm Hg or arterial oxygen saturation at 88% to 92% to avoid potential oxygen toxicity (see Table 3-3 in Chapter 3). Such adjustments may best be accomplished in the ICU rather than the ED, after the entire clinical scenario can be analyzed and all interventions are appropriately adjusted.



Positive End-Expiratory Pressure


PEEPe is typically set at 5 to 8 cm H2O. Most patients should be started at a PEEP of 5 cm H2O, which is considered a physiologic level. It is used to offset the gradual loss of functional residual volume (FRC) in supine, mechanically ventilated patients. PEEP can be increased by 2 cm H2O every 10 to 15 minutes as needed or tolerated by patients who remain hypoxic. The initial goal is to reduce Fio2 to nontoxic levels. This goal is coming under increasing scrutiny as new information challenges the time frame and concept of O2-induced lung injury at Fio2 levels greater than 0.6.9 Exercise care when using PEEP levels higher than 8 cm H2O in the setting of elevated intracranial pressure (ICP),10 unilateral lung processes, hypotension, hypovolemia, or pulmonary embolism. High PEEP can potentially lead to hypotension as it increases intrathoracic pressure and decreases venous return and subsequently cardiac output.




Waveform


The waveform determines how the ventilator delivers the flow of gas. It is traditionally set to a “decelerating waveform” in an effort to optimize recruitment because of different time constants in the lung.





I/E Time Ratio


The ratio of inspiratory time (the time that it takes to take a breath) to expiratory time (the time that it takes to exhale a breath) is automatically reported in some modes, whereas in others it is dialed in.


The normal I/E ratio in a spontaneously breathing, nonintubated patient is 1 : 4.11 Intubated patients commonly achieve I/E ratios of 1 : 2. Shorter ratios may lead to decreased exhalation by compromising Te. In its extreme form, inverse ratio ventilation (IRV), the normal pattern of breathing is reversed. A longer time is spent in inhalation to allow more time for oxygenation and recruitment. The decrease in Te can lead to air trapping, elevated mean airway pressure, and rising Pco2. These problems lead to hypercapnia, respiratory acidosis, and PEEPi12 (Fig. 8-8).






Modes of Ventilation


Once some of the standard features are understood, the next step is determining the ventilator’s target. Most ventilators can be set to achieve spontaneous breathing, volume-targeted ventilation, pressure-targeted ventilation, or some combination. In volume-targeted ventilation, the ventilator is set to reach a determined volume regardless of the pressure required to do so. Pressure-targeted modes are set to reach a determined pressure regardless of the volume generated. Dual modes combine the benefits of both strategies (Fig. 8-9).





Volume-Cycled Ventilation


Volume-cycled ventilation (VCV) may also be termed volume-limited, volume-control, volume-assist, or volume-targeted ventilation. Volume-targeted modes are the most commonly used and the most familiar mode of MV in adults. As its name implies, “volume”—in this case Vt—is the ventilator’s targeted parameter. With this target, the ventilator seeks to deliver a preset amount of gas. The ventilator will generate the necessary driving pressure to reach this “target.” In addition to Vt, the clinician sets the desired respiratory rate, Fio2, and PEEP. It should be noted that other important aspects of the mechanical ventilator can be controlled in this setting, such as waveform (decelerating or square), I : E ratio, flow rate, trigger, and sensitivity.


The time of gas flow is determined by the set volume, flow rate, and waveform of gas delivery. When the set volume is reached, gas flow is terminated and expiration passively begins. An advantage is that VCV delivers a reliable volume, but it does not take into account dynamic changes in lung compliance, which may alter the ability of the lung to accept delivered gas in gas-exchanging alveoli.



Pressure-Cycled Ventilation


Pressure-cycled ventilation (PCV) may also be termed pressure-limited, pressure-control, pressure-assist, or pressure-targeted ventilation. As its name implies, “pressure” is the ventilator’s targeted parameter. The ventilator will generate an inspiratory pressure that has been set by the clinician. With this target, the ventilator alters gas flow to achieve and maintain a preset airway pressure for the duration of a preset inspiratory time (Ti). Gas flow is terminated when the preset pressure is achieved. The volume delivered is determined by the compliance of the patient’s respiratory system, airway resistance, Ti, and the pressure target. In addition, the clinician sets the desired PEEP, respiratory rate, Fio2, Ti, I : E ratio, and trigger mode. Pressure is maintained with a variable or intermittent flow rate for the set Ti. In the setting of hypoxemia, Ti may be increased quite precisely to increase mean airway pressure and oxygenation. This strategy is much more difficult, if not impossible, to manipulate with VCV.


An advantage of PCV is that airway pressure is tightly managed to limit or eliminate alveolar overdistention and to reduce ventilator-induced lung injury.14 It should be noted that the clinician does not control waveform or peak inspiratory flow. Patients can generate their desired flow rate and thus reduce air hunger. Pressure-targeted modes, which are growing in popularity, might have better pressure distribution, improved dissemination of airway pressure, and greater distribution of ventilation.14


One problem with PCV is that the volume received by the patient is potentially variable. Any change in system compliance or resistance (or both) will affect the Vt generated. For example, if the patient bites on the ET tube or a mucus plug develops, the set pressure that was generating an adequate volume will no longer do so. In contrast, a sudden increase in system compliance might result in the generation of Vt that may be considerably larger than desired. Instead of the traditional pressure alarm limits, one must adjust and be cognizant of Vt and minute ventilation alarm settings. Uncertainties such as these have led many clinicians to favor volume-targeted strategies or dual controlled strategies in the acute care setting.



Modes of Ventilation Commonly Used in the ED


Assist/control (AC) and synchronized intermittent mandatory ventilation (SIMV) are the ventilation modes most commonly used in the ED. Both are acceptable, and no data have demonstrated a better outcome with either mode. Other modes are also acceptable based on clinician preference.



Assist/Control Ventilation


Here, ventilator-initiated breaths, known as machine breaths (i.e., control breaths), are provided at a preset rate. Every breath is fully supported by the ventilator, regardless of whether the breath is initiated by the patient or the ventilator. The clinician sets the base ventilation rate, but if the patient tries to breathe faster than the set rate, additional breaths can be initiated by the patient, known as spontaneous breaths (i.e., assist breaths). A potential downside is inappropriate hyperventilation.


AC modes may be either volume or pressure cycled. Both assist breaths and control breaths will reach the set target, be it a set volume (in a volume-targeted mode) or a set pressure (in a pressure-targeted mode). In VCV, the spontaneous breath receives the same Vt that is set for the machine breath. In PCV, the spontaneous breath receives the same pressure that is set for the machine breath. To be more specific, in the volume-targeted mode, the clinician sets Vt, as well as the inspiratory flow rate, flow waveform, sensitivity to the patient’s respiratory effort (i.e., trigger), and the basal ventilatory rate. In the pressure-targeted mode, the clinician determines the basal ventilatory rate and how sensitive the ventilator will be to the patient’s respiratory effort and also selects pressure levels and Ti. Hence in this mode Vt is not set by the ventilator but is dependent on the compliance of the lung and chest wall and airway pressure. This helps avoid pressure-induced lung injury, but a specific Vt is not guaranteed.


For the patient to trigger the ventilator and initiate flow for a spontaneous breath, mean airway pressure must decrease by a preset amount below PEEP if set on a pressure trigger or flow to be generated if set on a flow trigger. The amount necessary to open the inflow valve is the sensitivity setting.


Caution should be exercised to avoid auto-PEEP (also known as breath stacking) when using volume-targeted AC modes. Because each mechanically delivered breath is given at full Vt, patients with a high actual respiratory rate on AC may not have sufficient time to completely exhale between breaths. This results in progressive air trapping, which leads to an increase in auto-PEEP (PEEPi) (see Fig. 8-6). This is of clinical concern in patients with asthma, in whom auto-PEEP can significantly reduce cardiac output and even promote cardiovascular collapse.



Synchronized Intermittent Mechanical Ventilation


SIMV provides breaths at a preset rate (machine breath), similar to the AC mode. The patient can initiate an additional spontaneous breath between the mandated or preset number of ventilator-supported breaths. Such spontaneous breaths above the preset ventilatory rate are not supported by the ventilator, and the patient receives only a spontaneous Vt that reflects the depth and time spent in the patient-controlled inspiration. For each of these nonmandatory (i.e., spontaneous) breaths, the patient has a high work of breathing. SIMV is typically partnered with PSV to aid in spontaneous breathing support and to overcome the intrinsic resistance associated with MV. This mode was initially recommended by those who thought that as a patient’s need for mechanical ventilatory support decreased, the set respiratory rate could be decreased and the patient “weaned” to PSV alone and ensuing extubation. Subsequent data have shown that this method of liberation actually increases the number of ventilator days.15 The synchronized version of intermittent MV allows the ventilator to attempt to coordinate spontaneous and machine breaths to prevent it from delivering a scheduled breath on top of a spontaneous breath or during exhalation after a spontaneous breath. This could lead to elevated mean airway pressure, alveolar overdistention, and biotrauma.16


Both volume-targeted ventilation and pressure-targeted ventilation modes can be set to either an AC or SIMV mode to achieve the desired minute ventilation. In a chemically paralyzed patient with no intrinsic respiratory drive, AC and SIMV look virtually identical. Both will reach their target (volume or pressure) at the set rate. If a patient triggers the ventilator at a rate greater than the set rate, these two strategies diverge. In AC, each breath above the set respiratory rate will result in a full mechanically supported breath to reach either the set volume or pressure target. In SIMV, the ventilator will give only the set number of breaths that the clinician has selected. Each additional breath will require the patient to generate a spontaneous Vt without mechanical assistance. This patient-generated breath must overcome any resistance caused by the artificial airway and ventilator circuitry. Pressure support should be added to SIMV for patient-generated breaths to reduce any increase in the work of breathing related to the resistance imposed by the ventilator circuit and ET tube.



Advanced Modes of Mechanical Ventilation



Dual Control Modes


Advanced modes of mechanical ventilation use a closed-loop ventilator logic that combines the features of volume- and pressure-targeted ventilation (Box 8-1). These modes automatically alter control variables, either breath to breath or within a breath, to ensure a minimum Vt or minute ventilation.17 Detailed explanation of these modes is beyond the scope of this chapter. If these modes are encountered, one should discuss options with a respiratory therapist and critical care medicine specialist.




Other Modes




Airway Pressure Release Ventilation and Bi-Level Ventilation


Both these modes are proprietary names yet function in essentially the same manner. The clinician sets a pressure high, a pressure low, and a time at each level (time high and time low). Although at first glance this appears to be similar to PCV, it differs markedly in that the majority of time is spent at pressure high with brief periods at pressure low. The patient typically spends 4 to 6 seconds in time high. Pressure high may be as high as 40 cm H2O or greater. Ventilation occurs during the release from pressure high to pressure low. Time low is typically 0.2 to 0.8 second in restrictive lung disease and 0.8 to 1.5 seconds in obstructive lung disease. It is probably prudent to start at 0.8 and titrate to meet individual patient requirements. Time low is also referred to as the release phase.19 The long time that high-level pressure is maintained achieves oxygenation, and the short release period achieves clearance of CO2 (Fig. 8-10). The long time at high-level pressure results in substantial recruitment of alveoli from markedly different regional time constants at rather low gas flow rates. The establishment of PEEPi by the short release time enhances oxygenation. CO2 clearance is aided by recruitment of the patient’s lung at close to total lung capacity. Elastic recoil creates large-volume gas flow during the release period.



In a paralyzed patient, airway pressure release ventilation and bilevel ventilation (APRV/Bi-Level) are identical to pressure-targeted IRV. For these reasons, some have described this mode as inverse ratio ventilation. A major difference between APRV/Bi-Level and IRV is that IRV typically requires chemical paralysis or heavy sedation. APRV/Bi-Level is a fundamentally different mode from cyclic ventilation. This mode allows the patient to spontaneously breathe during all phases of the cycle, thus making it relatively more comfortable and reducing the level of sedation or paralysis needed. This mode is enabled to succeed by having a floating valve that is responsive to the patient’s needs, regardless of the location within the respiratory cycle. The patient is allowed to breathe in or out during the pressure high phase and during the release phase. Accordingly, the sequence is called a phase cycle. There is no set Ti or Te and no readily identifiable respiratory rate in the traditional sense. During the pressure high phase, patients may exhale 50 to 200 mL or more of gas as the lung volume becomes full of gas. This is not a full exhalation, and the release of excess gas should not be counted as a breath. APRV has been used successfully for neonatal, pediatric, and adult forms of respiratory failure. It is considered an alternative open–lung model approach to MV.19


Given the spontaneous nature of the mode, there should be virtually no need for continuous infusion of neuromuscular blocking drugs in patients placed on this mode of ventilation. This may result in a shorter length of ICU stay and a reduced incidence of prolonged neuromuscular blockade syndrome. The need for sedatives is reduced because patients are more comfortable on this spontaneous mode than on cyclic ventilation.20 APRV/Bi-Level has gained popularity in patients with hypoxemic respiratory failure because it improves oxygenation by optimizing alveolar recruitment and image matching.21


A common mistake with this mode is setting time low too long. This essentially mimics a pressure-targeted SIMV strategy. Transport of patients on APRV with pressure high greater than 20 cm H2O should occur with the patient attached to the ventilator instead of being hand-ventilated.22 Hand ventilation is unable to match the manner of gas delivery and pressure dynamics that the patient requires. Attempts at hand ventilation, even with an appropriately set PEEP valve, are frequently complicated by unexpected hypoxemia and hemodynamic instability.

Only gold members can continue reading. Log In or Register to continue

Sep 17, 2016 | Posted by in SUGERY, ORTHOPEDICS & ANESTHESIA | Comments Off on Mechanical Ventilation
Premium Wordpress Themes by UFO Themes