Acs0806 Mechanical Ventilation

Acs0806 Mechanical Ventilation

• 1.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 1 6 MECHANICAL VENTILATION Matthew J. Sena, M.D., and Avery B. Nathens, M.D., Ph.D., M.P.H., F.A.C.S. Approach to the Use of the Mechanical Ventilator Ventilation and Oxygenation Patients requiring mechanical ventilation account for a large per- centage of admissions to medical and surgical intensive care units. An essential concept in mechanical ventilation is the distinction The initial indications for mechanical ventilation can be divided between two key processes, ventilation and oxygenation. The pri- into two main categories: (1) airway instability necessitating endo- mary purpose of ventilation is to excrete carbon dioxide. The tracheal intubation (as a consequence of operation, brain trauma, ˙ minute ventilation (VE) is the total amount of gas exhaled per or intoxication) and (2) primary respiratory failure from any of minute, computed as the product of the rate and the tidal volume several diverse causes, including the acute respiratory distress syn- (VT). Minute ventilation has two components, alveolar ventilation drome (ARDS), trauma, cardiogenic pulmonary edema, and ˙ ˙ (VA) and dead space ventilation (VD). Under normal conditions, exacerbation of chronic obstructive pulmonary disease (COPD).1 ˙ approximately two thirds of VE reaches the alveoli and takes part In the first category, ventilator management is relatively straight- ˙ in gas exchange (VA); the remaining third moves in and out of the forward, and support is temporary, maintained only until the ˙ conducting airways and nonperfused alveoli (VD). Thus, the ratio patient’s airway is stabilized. In the second category, a prolonged of dead space to tidal volume (VD/VT) is normally 0.33. The period of mechanical ventilation (> 2 to 3 days) is frequently amount of CO2 excreted is directly related to the amount of alve- required.The majority of ventilated ICU patients fall into this sec- olar ventilation and inversely proportional to the partial pressure ond group,1 and it is these patients in whom specific attention of CO2 in the alveoli (PACO2). During spontaneous breathing, should be paid to the cause of respiratory failure and the goals of ˙ VE is regulated by the brain stem respiratory center. The brain therapy. The ventilator mode and settings can then be appropri- stem respiratory center responds primarily to changes in plasma ately tailored to minimize lung injury and facilitate resolution of pH and in the partial pressure of CO2 in arterial blood (PaCO2). the underlying disease. In the face of normal CO2 production (~ 200 ml/min) and nor- Proper use of a mechanical ventilator requires a solid under- mal minute ventilation (6 L/min), alveolar ventilation amounts standing of normal and abnormal pulmonary mechanics, gas to approximately 4 L/min and corresponds to a PaCO2 of exchange, and the relation between systemic oxygen delivery and 40 mm Hg. consumption. Mechanical ventilators, along with currently avail- ˙ In a patient requiring mechanical ventilation,VE is at least par- able noninvasive and invasive monitoring devices, allow support tially determined by the mode and settings of the ventilator. of critically ill patients while the acute physiologic derangements Respiratory rate and tidal volume can be set independently, and that led to respiratory failure resolve. In addition, specific ventila- the mode of ventilation can be set to allow additional spontaneous tor strategies geared toward minimizing further lung injury and breathing if necessary. In most cases, the primary goal is mainte- expediting the process of liberation from the ventilator not only nance of a near-normal PaCO2. The physician must be cognizant yield improved support of patients with respiratory failure but also of factors that might increase CO2 production (e.g., fever, sepsis, appear to have an impact on outcome. ˙ injury, and overfeeding) or VD (e.g., lung injury, ARDS, and mas- Ventilator terminology has become increasingly complex as sive pulmonary embolism), any of which would increase the VE ˙ the technology has advanced, but the basic principles of man- requirements in a ventilated patient. agement remain unchanged: to facilitate gas exchange for tissue Oxygenation refers to the equilibrium between oxygen in the oxygen delivery, to provide ventilation for removal of carbon pulmonary capillary blood and oxygen in inflated alveoli. The dioxide, and to minimize the detrimental effects of both endo- oxygen tension gradient between the alveoli and the capillaries tracheal intubation and mechanical ventilation. With these pri- favors the transfer of oxygen into the blood. Although the partial orities in mind, the clinician can use an evidence-based pressure of oxygen in arterial blood (PaO2) is partially dependent approach to ventilator management as a component of multi- on ventilation, it depends less on adequate alveolar ventilation modal therapy to improve patient outcome in the ICU. Such than on the appropriate matching of pulmonary blood flow to management includes use of a lung-protective strategy for well-inflated alveoli, a process referred to as ventilation-perfusion patients with acute lung injury (ALI) or ARDS, performance of ˙ ˙ ˙ ˙ (V/Q ) matching. V/Q matching can be affected by many factors, daily spontaneous breathing trials (SBTs) to identify patients including patient position, airway pressure, pulmonary parenchy- who are ready for liberation from the ventilator, and, when pos- mal disease, and small-airway disease. The efficiency of V/Q ˙ ˙ sible, consideration of a nurse-driven or respiratory therapist– matching, and thus of oxygenation, can be evaluated by measur- driven protocol to minimize delays in extubation. Newer thera- ing the PaO2 at a known value of the fraction (concentration) of pies have been developed that offer attractive alternatives to con- inspired oxygen (FIO2). Under normal circumstances, oxygena- ventional modes of ventilation. Most such therapies are of tion is very efficient, with PaO2 values approaching 90% of PAO2. unproven efficacy, and must therefore be employed with caution Its efficiency can be assessed by calculating the alveolar-arterial in clinical settings. Nonetheless, they provide the clinician with oxygen gradient (i.e., PAO2 – PaO2). more options for treating patients with advanced respiratory fail- Under normal conditions, PaO2 is approximately 90 mm Hg.To ure and should be considered in extreme cases. determine PAO2, the following formula is employed:
• 2.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 2 Mechanical ventilation is initiated Initial ventilator settings are as follows (depending on clinical scenario): FIO2 = 0.5; PEEP = 5 cm H2O; respiratory rate = 12–15 breaths/min; VT = 8–10 ml/kg predicted body weight. Measure SaO2. SaO2 < 90% Increase FIO2 in stepwise manner to keep SaO2 ≥ 90%. Approach to Use of the Increase PEEP by 2–5 cm H2O. Mechanical Ventilator Continue increasing PEEP by 2–5 cm H2O to maximum of 20 cm H2O if SaO2 < 90% despite FIO2 ≥ 0.8. Identify and treat cause of respiratory failure. Look for evidence of acute lung injury. Evidence of ALI is present Utilize low–tidal volume (lung-protective) ventilation: • Reduce VT to 6 ml/kg. • Increase RR to up to 35 breaths/min to achieve pH > 7.20 and PaCO2 ~ 40–50 mm Hg. = (Traumatic brain injury is a relative contraindication to this approach. Patients without intracranial hemorrhage but with intracranial pressure monitors may be considered if PaCO2 is normal and SaO2 > 95%.) Attempt to determine best PEEP through clinical or invasive assessment of DO2. Measure SaO2. SaO2 < 90% SaO2 ≥ 90% Diagnose and treat associated conditions: Measure Pstat. • Pneumothorax • Hydrothorax/hemothorax • Asynchrony (increase sedation; consider NMBA) Consider adjunctive measures: • Nitric oxide Pstat > 30 cm H2O Pstat ≤ 30 cm H2O • Prone positioning • HFOV Reduce VT in stepwise manner to 4 ml/kg to keep • ECLS Pstat ≤ 30 cm H2O. (In patients with morbid obesity or ascites, Pstat may reflect transdiaphragmatic pressure rather than transpulmonary pressure. The lung-protective approach should be maintained, but consideration should be given to allowing a higher Pstat before lowering VT significantly below 6 ml/kg.) When lung compliance improves, begin increasing VT to 6 ml/kg while maintaining Pstat ≤ 30 cm H2O. Continue lung-protective ventilation strategy until PaO2/FIO2 ratio ≥ 300 or patient meets criteria for SBT.
• 3.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 3 SaO2 ≥ 90% ~ Adjust RR to keep PaCO2 = 35–45 mm Hg, unless severe bronchospasm or COPD is present. Evidence of ALI is absent Continue support until gas exchange improves. As hypoxemia resolves, • Reduce FIO2 as tolerated: ≤ 0.5 to keep SaO2 ≥ 90% • Reduce PEEP to ≤ 8 cm H2O (in steps of 2–5 cm H2O). Perform daily assessment for liberation from ventilator. Determine whether patient meets criteria for SBT. Patient passes SBT Patient fails SBT Assess stability of airway. Determine cause of failure and attempt to correct it. Resume completely supported ventilation for 24 hr, then reattempt SBT. Patient passes SBT on Patient persistently fails SBT subsequent attempt Consider tracheostomy. Assess stability of airway. Airway is stable Airway is unstable Extubate patient. Consider tracheostomy. Resume daily SBTs with CPAP or tracheostomy collar. For patients with prolonged ventilator dependence (≥ 2 wk), consider • Planned, gradual reduction of pressure support (PSV wean) or • Planned, gradual increases in duration of SBT (2–12 hr) until patient’s endurance improves.
• 4.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 4 PAO2 = [FIO2(barometric pressure –PH2O) – PaCO2/RQ] through an inspiratory effort, which leads to the delivery of a breath at a variable flow rate to meet a preset pressure. As the lung where RQ represents the respiratory quotient and PH2O represents inflates, compliance decreases and flow decreases to maintain a the partial pressure of water vapor at sea level. Normally, at sea constant inspiratory pressure. The result is a descending flow level, barometric pressure is approximately 760 mm Hg, FIO2 is curve that is similar to air flow in unassisted breathing. Cycling in 0.21, PH2O is 47 mm Hg, PaCO2 is 40 mm Hg, and RQ is 0.8. this mode occurs when flow declines to a specified percentage of Accordingly, the maximal flow rate (approximately 5% of the peak flow rate in PAO2 = [0.21(760 – 47) – 40/0.8] some ventilator models) [see Figure 1]. When inspiratory flow PAO2 ≅ 150 – 50 ceases, the patient exhales passively. PAO2 ≅ 100 Pressure-controlled ventilation is related to PSV in that flow PAO2 – PaO2 ≅ 100 – 90 ≅ 10 descends in amplitude during the inspiratory cycle. It differs from PSV primarily in that the inspiratory time is set by the ventilator, Thus, the alveolar-arterial gradient under these conditions is ap- not by the patient. PCV is generally used in the assist-control proximately 10 mm Hg, which falls within the standard range of mode, which allows full support of patient-initiated breaths in 8 to 12 mm Hg. addition to ventilator-initiated breaths (which are time triggered). A shorthand method of quantifying the degree of hypoxemia is It can also be used in conjunction with the intermittent mandato- to calculate the PaO2/FIO2 ratio (also referred to as the P/F ratio), ry ventilation (IMV) mode in newer ventilator models.The largest which is simply an assessment of the efficiency of gas exchange. drawback of PCV is that VT can change as lung compliance At sea level, with the patient breathing room air, P/F ≅ 100/0.21 changes, necessitating frequent adjustments to ensure adequate ≅ 500. ˙ VE. For example, as the lung becomes less compliant with increas- ˙ Unlike regulation of VE, adjustment of the rate or VT generally ˙ ing pulmonary edema, VT will decrease, as will VE. This problem has little effect on PaO2, except at extremely low levels of ventila- can be addressed by using another form of pressure- tion. Greater effects on arterial oxygenation are achieved through limited ventilation, known as pressure-regulated volume control adjustment of either FIO2 or mean airway pressure (Paw), both of (PRVC) [see Combined Modes of Ventilation, below]. which can be readily manipulated with a mechanical ventilator. Volume-controlled ventilation (VCV) is delivered at a set fre- quency (in the IMV or the assist-control mode) or may be patient Ventilator Modes initiated (in the assist-control mode). After the ventilator is trig- gered, a fixed flow of gas is generated for a specific time, thus pro- Current mechanical ventilators possess a wide, and potentially viding a preset inspiratory volume (volume = flow × time). confusing, array of modes, settings, and capabilities. All of them, Volume-limited ventilation is generally easier to regulate than pres- however, control three variables: trigger, limit, and cycle. sure-limited ventilation, but it may be less comfortable for awake The trigger variable is the signal that serves to initiate the inspi- patients, because the flow curve is a square wave, which is marked- ratory phase. This signal occurs as a result of patient effort that ly different from the normal inspiratory flow pattern in nonventi- leads to a change in either flow or pressure within the ventilator lated patients. circuit. Flow-triggered ventilators deliver a continuous flow of gas across the inspiratory and expiratory limbs of the ventilator circuit MANDATORY VERSUS SPONTANEOUS VENTILATION and initiate the inspiratory phase when patient effort results in a There are several modes of ventilation that provide mandatory change in this flow. The required change can be as little as 0.1 ventilator support with or without patient-triggered ventilation. L/min; the sensitivity of the trigger is decreased by increasing the For example, assist-control ventilation ensures delivery of a mini- required flow change and therefore increasing the patient effort ˙ mum (mandatory) set VE but also allows additional patient-trig- necessary to begin inspiration. Pressure-triggered ventilators initi- gered (spontaneous) breaths. Each breath, regardless of the trig- ate the inspiratory phase when a patient’s spontaneous effort ger, is completely supported, so that either a fixed volume (with results in a change in pressure. At the most sensitive setting, a VCV) or a fixed pressure (with PCV) is provided for a preset pressure change of approximately –1 cm H2O is required; at the inspiratory time. Full support can be provided by increasing the least sensitive setting, a change of –15 cm H2O is required. ˙ mandatory rate. If the patient has respiratory drive, VE might be Finally, a time trigger is used to start the inspiratory phase in increased by adding spontaneous breaths.The major drawback of mandatory ventilation modes, as well as in assisted modes. this approach is that agitated patients may become hyperventilat- The limit variable is the maximal set inspiratory pressure or ed and may manifest respiratory alkalosis if not sedated. flow. Pressure-controlled ventilation (PCV) and pressure-support The IMV mode allows only the preset number of breaths to be ventilation (PSV) are both modes of pressure-limited ventilation. supported. In most cases, breaths are synchronized with the pa- Because volume is the product of flow and time, volume-con- tient’s effort (so-called synchronized IMV [SIMV]) if sponta- trolled ventilation is actually flow-limited ventilation during the neous respiration is occurring. Patient efforts at inspiration above inspiratory phase with the inspiratory time set independently. the preset frequency are not supported with gas flow from the The cycling variable is the factor that terminates the inspirato- ventilator unless pressure support is added. In the past, IMV was ry cycle (i.e., time, flow, pressure, or volume). To add more con- frequently used as a weaning mode: by gradually decreasing the fusion, each breath can be considered as a mandatory breath set frequency, patients gradually assumed a greater role in their (which is time triggered), a spontaneous breath (which is patient own respirations.Today, however, it is not frequently employed for initiated), or a combination of the two. this purpose and plays only a limited role in weaning patients from PRESSURE-LIMITED VERSUS VOLUME-LIMITED VENTILATION the ventilator. Pressure-support ventilation is the simplest form of pressure- INSPIRATORY TIME, FLOW, RESPIRATORY RATE, AND limited ventilation and, by definition, is a purely spontaneous INSPIRATORY-EXPIRATORY RATIO mode of ventilatory support. In PSV, the patient triggers a breath Inspiratory time, flow, respiratory rate, and inspiratory-expira-
• 5.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 5 Volume-Limited (Flow-Limited) Ventilation Pressure-Limited Ventilation Volume-Controlled Volume-Controlled Pressure-Controlled (Assist Control, Time Trigger) (Assist Control, Patient Trigger) (Assist Control, Time Trigger) Pressure-Support 1,200 Square Flow Wave, Decelerating Flow Wave, Expiratory 800 Variable Pressure Constant Pressure Phase Flow (ml/sec) 400 Begins 0 –400 –800 Spontaneous Effort –1,200 30 Pressure (cm H2O) 25 Ventilator Trigger Set to Inspiratory Time Determined –2 cm H2O by Patient = 1.2 sec 20 PS = 10 cm H2O 15 10 Pressure 5 Deflection of –2 cm H2O 0 Tidal Volume (ml) 1,000 750 VT = 600 ml 500 250 0 0 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 9 10 Time (sec) Time (sec) Time Patient (pressure in this case, but Time Patient (pressure in this case, •RR set at 15 breaths/min may be flow) •RR set at 15 breaths/min but may be flow) Trigger variable •Inspiratory time = 1 sec; thus, •Sensitivity set at –2 cm H2O •Inspiratory time = 1 sec; thus, •Sensitivity set at –2 cm H2O I/E = 1:3 I/E = 1:3 Limit Flow Flow Pressure Pressure variable •Square wave, 600 ml/sec •Square wave, 600 ml/sec •∆P = 10 cm H2O above •∆P = 10 cm H2O above PEEP PEEP Volume or time Volume or time Time Flow (determined by patient’s Cycling •Volume = flow × time; thus, •Volume = flow × time; thus, lung compliance) variable 600 ml = 600 ml/sec × 1 sec 600 ml = 600 ml/sec × 1 sec •Gas flow ceases when flow rate inspiratory time inspiratory time reaches 5% of peak value Figure 1 Shown are flow, pressure, and volume profiles in volume-limited and pressure-limited ventilation modes. AIRWAY PRESSURES, LUNG INJURY, AND OXYGENATION tory (I/E) ratio are all closely related. In most patients, they can be preset to mimic the normal respiratory cycle.The normal I/E ratio The flow of gas through the ventilator circuit produces pressure is approximately 1:3. With the respiratory rate set at 15 breaths/ both at the level of the endotracheal tube and across the alveolar min, the inspiratory time is 1 second, with 3 seconds of expirato- surface. These pressures, though related, have different implica- ry time. The flow rate can then be manipulated so as to achieve a tions for the assessment and treatment of pulmonary dysfunction desired tidal volume (in VCV) or can be adjusted automatically by [see Figure 2]. Peak inspiratory pressure (PIP) is the pressure mea- the ventilator so as to achieve a certain pressure (in PCV). sured in the ventilator circuit during maximal gas flow; it primar- Manipulation of these variables can be useful in certain condi- ily represents the interaction between the inspiratory flow rate and tions, such as a high level of intrinsic positive end-expiratory pres- airway resistance. Mean airway pressure is the area under the sure (PEEP) [see Discussion, Special Problems in Ventilator Man- pressure-time curve divided by the time required for a complete agement, Chronic Obstructive Pulmonary Disease, below]. In the respiratory cycle. Because the normal respiratory cycle is domi- past, manipulation of the I/E ratio has been used to treat severe nated by the expiratory phase, Paw is determined primarily by hypoxemia [see Alternative Modes of Ventilation and Adjunctive PEEP. Paw is important in that it has a direct effect on alveolar Therapies, Inverse-Ratio Ventilation, below]. recruitment and gas exchange; it also is the major determinant of
• 6.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 6 30 adjusted accordingly. Certain patients—specifically, those with PIP profound hypoxemia—may require an immediate increase in FIO2, 25 PEEP, or both; such increases should not be postponed to await Pressure (cm H2O) Pstat 20 blood gas results. FIO2 and PEEP can be manipulated on the basis of information from pulse oximetry, so that there is less need for 15 frequent arterial blood gas assessment. Further treatment should be prioritized on the basis of the underlying problem of oxygena- 10 PEEP tion or ventilation. 5 3 OXYGENATION The purpose of ensuring adequate arterial oxygenation is ulti- 0 1 2 3 4 5 6 7 mately to maintain adequate delivery of oxygen to the tissues. A Time (sec) PaO2 higher than 60 mm Hg generally results in an arterial hemo- Figure 2 Shown are measured ventilator pressures during a sin- globin saturation (SaO2) of 90% or greater and is sufficient for gle machine-triggered volume breath during VCV. (PEEP—posi- most patients, provided that the other components of oxygen tive end-expiratory pressure; PIP—peak inspiratory pressure; delivery are normal or nearly so. An adequate PaO2 can be ob- (Pstat—static pressure measured during a 0.5 sec inspiratory pause) tained by altering either FIO2 or Paw. Increasing the FIO2 is the sim- plest maneuver, but it is not necessarily the correct adjustment in intrathoracic pressure and thus is the parameter to follow when patients with pulmonary dysfunction. Nonetheless, it should be there is concern about the cardiovascular sequelae of higher ven- tried first, while other options are being considered. Generally, a tilatory pressures. Static pressure (Pstat) is measured in the venti- moderate increase in FIO2 (to ≤ 0.6) has minimal adverse conse- lator circuit during a 1-second pause at the end of inspiration. Pstat quences. The desired and immediate effect is to increase the gra- is generally considered to be the pressure distending the alveoli, dient for oxygen diffusion across the alveolar and pulmonary cap- on the assumption that intrathoracic pressure is equivalent to illary membranes. In normal lungs, this increased gradient results atmospheric pressure; this distending pressure is also referred to in a proportional increase in PaO2. If an intrapulmonary shunt is as transpulmonary pressure. Limitation of Pstat plays an important present, however, increasing FIO2 has little effect on PaO2. role in minimizing ventilator-induced lung injury [see Discussion, Often, an effort to increase FIO2 is combined with some mea- Special Problems in Ventilator Management, Acute Lung Injury sure aimed at increasing Paw, the idea being that a smaller increase and Acute Respiratory Distress Syndrome, below]. in FIO2 will then be required to produce the desired effect. Although a high FIO2 (> 0.6) has few negative consequences in COMBINED MODES OF VENTILATION the short term, prolonged maintenance of FIO2 at this level can Many newer modes of ventilation are hybrids, incorporating combinations of pressure control and volume control and combi- nations of spontaneous and mandatory breathing. One such 30 mode is pressure-regulated volume control. PRVC is a variant of Effective PCV but has the ability to prevent significant changes in VT if Compliance lung compliance changes. The ventilator accomplishes this by 20 continuously evaluating changes in delivered VT at a given pres- 200 sure over several respiratory cycles and adjusting the pressure accordingly.The operator can preset the inspiratory pressure limit PaO2 to prevent barotrauma. 40 Another hybrid mode is mandatory minute ventilation, which is a form of assisted VCV. In the assist mode, the patient may trig- Cardiac ger a complete volume-cycled breath.The operator sets a specific Output ˙ VE, rather than a specific rate and VT. In this way, the operator can O2 ensure adequate CO2 removal in patients with variable respirato- Delivery ry drive. In general, these combined modes are of limited utility in ven- Arterial Content tilator management. The overwhelming majority of patients can be managed with simple PCV or VCV or, if they have adequate respiratory drive, with PSV. Venous O2 Content Use of the Mechanical Ventilator in Respiratory Failure 0 5 10 15 20 25 After endotracheal intuba- PEEP (cm H2O) tion, the initial ventilator set- Figure 3 Shown are measurements for determining the optimal tings should be determined by PEEP The objective is to maximize the ratio of oxygen delivery to . assessing the cause and sever- oxygen consumption, which is determined by measuring mixed ity of the patient’s respiratory venous saturation. Here, the best PEEP is 10 cm H2O, even failure. After an initial stabilization period of approximately 30 though the PaO2 and the arterial oxygen content are higher at minutes, blood gas values should be obtained and the ventilator higher levels of PEEP.
• 7.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 7 6 ry to ry ra to pi Figure 4 Depicted is a static volume-pressure curve. A → A′ ira Ex 5 sp represents normal tidal ventilation in a normal lung, with a In A′ PEEP of 5 cm H2O and a tidal volume of 10 ml/kg × 70 kg = 4.5 700 ml. Cstat = ∆V/∆P = 700 ml ÷ 5 cm H2O/kg = 2 ml/cm H2O ÷ ∆V 70 kg. B → B′ represents normal tidal ventilation in a lung 4 3.8 A from an ARDS patient, with a PEEP of 5 cm H2O and a tidal Normal ∆P volume of 10 ml/kg × 70 kg = 700 ml. Cstat = 700 ÷ 22 cm H2O ÷ Volume (L) FRC 70 kg = 0.45 ml/cm H2O/kg. C → C′ represents a low–tidal Overdistention 3 volume, lung-protective ventilation strategy, with a PEEP of 5 2.7 B′ cm H2O and a tidal volume of 6 ml/kg × 70 kg = 420 ml. Cstat = ry 420 ÷ 17 cm H2O ÷ 70 kg = 0.35 ml/cm H2O/kg. D → D′ repre- to sents a low–tidal volume, lung-protective ventilation strategy, pira B C ′,D′ D 2 Ex with a PEEP of 12 cm H2O (best PEEP), which is above Pflex. The tidal volume is 6 ml/kg × 70 kg = 420 ml. Cstat = 420 ÷ 10 al C rm ry cm H2O ÷ 70 kg = 0.60 ml/cm H2O/kg. The low–tidal volume to No ira strategy prevents the overdistention that occurs at higher DS Lower Inflection Point sp 1 AR In (Pflex, Derecruitment) tidal volumes, whereas the higher PEEP level prevents the derecruitment that can occur at lower volumes. Although PEEP Set at 12 cm H2O low–tidal volume ventilation strategies have been demonstrat- (Pflex + 2 cm H2O) ed to improve outcome in ARDS, higher PEEP levels, though 0 0 5 10 15 20 25 30 35 40 theoretically attractive, have not been shown to be efficacious. 12 22 Pressure (cm H2O) result in nitrogen washout, resorption atelectasis, and an increased referred to as best PEEP and generally corresponds to the PEEP pulmonary shunt fraction, with consequent exacerbation of setting that results in maximal static lung compliance.5 PEEP lev- hypoxemia. els above this point may increase arterial oxygen content (CaO2) The simplest way of increasing Paw is to increase PEEP. The but typically impair cardiac output and result in decreased DO2 ˙ purpose of PEEP is to prevent loss of functional residual capacity [see Figure 3]. (FRC), defined as the volume maintained in the lungs at the end The improvement in lung compliance achieved by increasing of expiration. In normal persons, FRC is maintained by a balance PEEP levels in patients with pulmonary dysfunction can be dem- between the negative intrapleural pressure and the elastic recoil of onstrated by using a static or low-flow volume-pressure curve to the lung. Negative intrapleural pressure is affected by patient posi- determine the lower inflection point (Pflex) [see Figure 4]. Pflex cor- tion. The upright position yields a greater negative intrapleural responds to the point at which recruitable alveoli open and pressure than the supine position, because of the weight of the become available for tidal ventilation and gas exchange. Setting abdominal viscera, which literally pull down on the diaphragm. In PEEP slightly higher (+2 cm H2O) allows tidal ventilation over the supine position, this pull is absent, and FRC may be as much the range of maximal lung compliance and prevents the repetitive as 25% lower in normal persons in the supine position.2,3 This end-expiratory derecruitment that may be associated with venti- decrease is even more pronounced in patients with ascites or lator-induced lung injury.6 Although this approach is theoretical- abdominal distention and especially in patients with intra-abdom- ly attractive, determining Pflex is technically cumbersome and may inal hypertension.4 In a normal alert patient, the loss of FRC can not be tolerated by the sickest patients. be reversed with changes in position or intermittent sigh breaths In practice, FIO2 and PEEP are manipulated simultaneously, throughout the respiratory cycle, and there will be minimal net with the specifics depending on the cause of arterial hypoxemia impact on pulmonary physiology. In a supine ventilated patient, and on an empirical determination of best PEEP. Best PEEP is the loss of FRC results in progressive atelectasis, intrapulmonary determined by means of stepwise increases in PEEP coupled with shunting, and hypoxemia. Accordingly, small amounts of PEEP serial assessments of arterial oxygenation and cardiovascular func- (5–10 cm H2O) should be delivered to help restore FRC to levels tion. At high levels of PEEP (> 15 cm H2O), consideration should adequate for maintaining normal gas exchange and preventing be given to using a pulmonary arterial catheter for assessment of hypoxemia in patients without significant pulmonary dysfunction. ˙ DO2 and mixed venous oxygen saturation (SvO2). If an increase in In patients with pulmonary dysfunction, FRC is lost through ˙ PEEP results in a drop in either SvO2 or DO2 , then the increase alveolar collapse. This collapse occurs by several means—for was detrimental to the goal of improving tissue oxygenation. As a example, through extrinsic pressure (see above) or through some rule, the maximal PEEP level that may provide benefit rarely combination of blood, pus, or secretions that results in occlusion exceeds 20 cm H2O.7 Above this level, it is fairly common for of small airways. In these settings, increasing PEEP improves V/Q˙ ˙ detrimental effects on cardiovascular function to predominate matching by “recruiting” these collapsed alveoli and thereby ˙ and DO2 to decline.8 bringing about improved gas exchange and PaO2. Unfortunately, The initial settings for FIO2 and PEEP depend on the clinical the increased intrathoracic pressure that may develop when PEEP scenario. Patients intubated for postoperative airway protection is increased can have detrimental effects on cardiac output. Be- may require an FIO2 of 0.3 and a PEEP of 5 cm H2O. In contrast, cause the overall goal is to improve tissue oxygen delivery (not just multiply injured patients who have been resuscitated may require PaO2), assessment of the net effect of an increase in PEEP should an FIO2 of 1.0 and a PEEP of 15 cm H2O. In either case, early ˙ take into consideration the effect on oxygen delivery (DO2 ). The evaluation of arterial blood gas concentrations can guide further ˙ amount of PEEP necessary to maximize DO2 in a given patient is manipulations. Alternatively, oxygen saturation and end-tidal CO2
• 8.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 8 values may be followed noninvasively with pulse oximetry and capnography. Sudden decreases in PaO2 (or SaO2) should be treat- Table 1—Criteria for Liberation from ed by first increasing FIO2 and then increasing PEEP.These mea- Mechanical Ventilation sures should be followed by attempts to ascertain the cause of the acute change. A first assessment—including suctioning, arterial Patient criteria for spontaneous breathing trial (SBT) to assess blood gas measurements, and a chest radiograph—should be car- readiness for liberation from mechanical ventilator ried out immediately, with particular attention to rapidly revers- Resolution or stabilization of underlying disease process ible causes (e.g., mucous plugging, pneumothorax, a large hemo- No evidence of residual pharmacologic neuromuscular blockade Spontaneous respiratory efforts thorax or hydrothorax, and cardiogenic pulmonary edema). A Hemodynamic stability (no recent increase in pressor or inotrope more detailed assessment should then follow to look for other requirements) possible causes (e.g., pulmonary embolism, worsening ARDS, Ventilator settings as follows: aspiration pneumonitis, and pneumonia). FIO2 ≤ 0.5 It must be kept in mind that in many cases, the etiology is mul- PEEP ≤ 8 cm H2O tifactorial, and that in the treatment of profound hypoxemia, it is PaO2 > 75 mm Hg important to address any and all correctable abnormalities, even Minute ventilation < 15 L/min when the potential gain is small. For example, in a patient with no pH 7.30 – 7.50 pulmonary reserve, drainage of a large hydrothorax may yield a Patient criteria to assess readiness for extubation significant improvement in oxygenation, whereas in a patient with Suctioning required less often than every 4 hr near-normal pulmonary function, this measure would have little, Good spontaneous cough if any, effect. Endotracheal tube cuff leak* In a minority of cases, a mismatch between patient effort and No recent upper airway obstruction or stridor† ventilatory support can result in increased work of breathing, pro- No recent reintubation for bronchial hygiene gressive respiratory muscle fatigue, and, on rare occasions, arteri- Criteria for a failed SBT‡ al desaturation. This situation, referred to as patient-ventilator Respiratory rate > 35 breaths/min for ≥ 5 min asynchrony, occurs as a consequence of the ventilator’s failure to SaO2 < 90% for ≥ 30 sec match the patient’s respiratory drive and pulmonary mechanics. It HR > 140 beats/min, or 20% increase or decrease from baseline only occurs during spontaneous breathing modes and may be sec- Systolic BP > 180 mm Hg or < 90 mm Hg ondary to a problem in the inspiratory trigger (inspiratory asyn- Sustained evidence of increased work of breathing (e.g., retrac- tions, accessory muscle use) chrony), the expiratory trigger (expiratory asynchrony), or the flow Cardiac instability or dysrhythmias rate (flow asynchrony).9 In these cases, direct patient observation pH ≤ 7.32 often suffices to establish the diagnosis. Therapy is aimed at improving the patient-ventilator interaction and may involve *Absence of a cuff leak is not an absolute contraindication to extubation. Each changing the mode of ventilation (e.g., from VCV or PCV to patient’s risk for postextubation upper airway obstruction should be assessed individually. PSV), the trigger setting (e.g., from pressure to flow), the inspira- †If extubation has recently failed because of airway obstruction, patient should be tory gas flow (either the rate or the waveform), or the cycling vari- assessed and the underlying cause addressed (if possible) before extubation is reat- able (time, flow, or pressure). Alternatively, in severe cases that tempted. Appropriate adjunctive measures (e.g., racemic epinephrine or helium- oxygen) should be available before patient is extubated. result in hypoxemia, it may be necessary to increase sedation to the ‡If any of these criteria are met, SBT should be terminated and the patient placed point where spontaneous respiratory efforts are eliminated. Neu- back on previous ventilator settings for 24 hr. romuscular blocking agents should only be considered if other measures have failed and hypoxemia is worsening. Generally, use of these agents should be avoided in critically ill patients when Currently, however, it is considered to be best suited for patients possible, because their administration has been associated with in whom the acute physiologic derangements leading to respira- significant complications.10 tory failure are resolving but who are not ready to be liberated from the ventilator.The main difference from past applications of VENTILATION PSV is that in current practice, pressure support is not intention- Adequate CO2 elimination can be achieved with either PCV or ally decreased over time [see Liberation from Mechanical Venti- VCV; the two methods can be used to achieve the same end lation, below]; instead, the patient is completely supported while points, and neither has any overwhelming advantage over the daily assessments of the patient’s readiness for extubation are other. Accordingly, it is reasonable to choose between them on the made. A potential advantage of PSV is enhanced patient comfort. basis of individual or institutional experience, simply for ease of This is a particularly important consideration when the patient’s management. The initial settings should include a VT of 8 to 10 overall condition is improving and minimization of sedation may ml/kg predicted body weight11 and a set respiratory rate of 12 to allow earlier extubation. 15 breaths/min. In general, the assist mode is preferable, because In the pressure-support mode, the patient determines the res- it allows the patient to regulate PaCO2 while still receiving com- ˙ piratory rate, inspiratory time, and VE. In a patient with normal pletely supported ventilation. If PCV is preferred, the inspiratory lung compliance, pressure support of between 5 and 10 cm H2O pressure can be adjusted at the bedside to achieve a VT of 8 to 10 is usually sufficient and should result in a VT of at least 600 ml. ml/kg. In either case, if the VD/VT ratio is nearly normal, the resul- Pressure support may be increased as needed for patient comfort tant alveolar ventilation should be sufficient to eliminate all of the and should be titrated to keep the respiratory rate below 25 metabolically produced CO2 and maintain a PaCO2 of 40 mm Hg. breaths/min. When pressures higher than 20 cm H2O are After approximately 30 minutes, blood gases should be measured required, most physicians elect to support the patient in the assist and the respiratory rate adjusted accordingly. mode until pulmonary compliance improves. In a patient with adequate respiratory drive, PSV is a reason- ˙ If PaCO2 is elevated despite a VE of 100 ml • kg-1 • min-1 (6–7 able choice. In the past, it was largely considered a weaning mode. L/min), then the metabolic production of CO2 is excessively high,
• 9.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 9 alveolar ventilation is an inappropriately low percentage of VT algorithm that incorporates clinical data, ventilator-derived data, (increased VD/VT), or both. To increase ventilation, the first step and laboratory data to determine the timing and likelihood of suc- should be to increase the respiratory rate in a stepwise fashion to cessful liberation from the ventilator. 20 to 25 breaths/min. In addition, the use of low-compliance ven- A randomized trial from 1995 compared four different wean- tilator tubing should be considered to minimize dead space in the ing methods: (1) daily 2-hour SBTs, (2) twice-daily 2-hour SBTs, ventilator circuit. If the PaCO2 is still elevated, VD and total CO2 (3) gradual reduction of pressure support, and (4) gradual reduc- production can be measured directly by using a metabolic cart. If tion of the IMV rate.The two SBT methods were superior in pre- CO2 production is higher than normal (130 ml • m-2 • min-1), it dicting successful extubation.13 In a follow-up study, the investi- can be decreased by reducing muscular activity or seizures, con- gators reported that a 30-minute SBT was as effective as a 2-hour trolling hypermetabolic states (if possible), and minimizing the SBT, and the shorter duration is now preferred for most pa- exogenous carbohydrate load. tients.14 Other traditional parameters used to predict the end of If the PaCO2 is still elevated after all of these measures have been the need for mechanical ventilation are respiratory rate, rapid- taken, the respiratory rate may be increased to 30 breaths/min or shallow breathing index (RSBI; calculated as frequency divided higher. It should be kept in mind, however, that the efficiency of by VT), VT, vital capacity, pressure-time product, and negative ventilation decreases as the respiratory rate increases. This loss of inspiratory force.These variables are useful adjuncts in the assess- ˙ efficiency occurs because the percentage of VE used for gas ment of a patient’s readiness to undergo an SBT, but they are not exchange decreases as the respiratory rate increases because in the highly accurate in predicting the likelihood of successful extuba- face of a fixed volume of dead space, there may be inadequate tion when used alone.15 time for alveolar emptying during expiration. In addition, steps It is not necessary for the acute process to have resolved com- may be taken to increase VT, though significant increases may pletely before an SBT can be performed, provided that other pre- worsen ventilator-induced lung injury through either barotrauma determined criteria are met [see Table 1].To await normalization of or volutrauma [see Discussion, Special Problems in Ventilator the P/F ratio or resolution of the chest x-ray abnormality would Management, Acute Lung Injury and Acute Respiratory Distress result in a needless delay in extubation should the patient success- Syndrome, below]. fully complete the SBT. Unnecessary prolongation of mechanical The importance of maintaining a normal PaCO2 is often over- ventilation heightens the risk of ventilator-associated pneumonia stated. Allowing PaCO2 to climb above 40 mm Hg has no intrinsic [see 7:17 Postoperative andVentilator-Associated Pneumonia], increas- detrimental effects in the absence of increased intracranial pres- es sedation requirements, postpones mobilization, and delays dis- sure, and it is not uncommon to permit the PaCO2 to rise to avert charge from the ICU.16 the adverse consequences of high tidal volumes and the ventilato- ry pressures required to generate these volumes. This approach, SPONTANEOUS BREATHING TRIAL referred to as permissive hypercapnia, is safe as long as pH re- SBTs should be performed on a daily basis once the acute res- mains above 7.15. Over time, pH will increase with compensato- piratory process has resolved and the patient is hemodynamically ry increases in renal bicarbonate preservation, provided that renal normal [see Table 1]. PEEP should be set at 5 cm H2O, with or with- function is normal. out an additional 5 cm H2O of pressure support (if the endotra- cheal tube is less than 7.0 mm in diameter).The FIO2 should be set to a value approximately 10% greater than required while the Liberation from patient is fully supported, and the patient should be allowed to Mechanical Ventilation breathe spontaneously for 30 to 120 minutes. At the conclusion of As the patient’s condition the SBT, blood gas values should be obtained, and the patient improves, it is useful to distin- should be placed back on the ventilator at the previous settings. At guish between the need for all times, the patient’s vital signs should be monitored for evidence continued endotracheal intu- of increased work of breathing. Criteria for a failed SBT include (1) bation and the ongoing re- significant changes in the respiratory rate, (2) evidence of increased quirement for mechanical work of breathing, (3) significant dysrhythmia, and (4) hemody- ventilation. The need for en- namic instability [see Table 1]. In addition, the arterial blood gas val- dotracheal intubation requires an assessment of airway stability ues should be evaluated for evidence of worsening hypoxemia or and is relatively straightforward (see below). Deciding when pul- hypercarbia, though it should be kept in mind that a normal PaCO2 monary function and respiratory muscle reserve are adequate for is less important if the pH is within the normal range. unassisted breathing is considerably more complex. The latter is If the patient successfully completes the SBT, extubation should what has traditionally been referred to as weaning from mechan- be attempted. There are few reasons to continue mechanical ven- ical ventilation. tilation in this situation. If it is suggested that the patient remain The term weaning implies a planned, gradual reduction in ven- intubated, one should ask whether a further delay in extubation tilator support whereby the patient assumes more and more of the would improve the chances of success, and if so, how. Among the work of breathing that had been performed by the ventilator.This reasons frequently cited for continuing intubation are altered men- is an inaccurate description of the actual process, which essential- tal status and inability to protect the airway, a potentially techni- ly involves assessment of a patient’s ability to sustain independent cally difficult reintubation, the presence of an unstable injury to ventilation and adequate gas exchange. Although the amount of the cervical spine, the likelihood of return trips to the operating support provided does decline as the patient improves, this decline room, and the need for frequent suctioning. Each of these reasons is patient driven and is different from the physician-driven reduc- must be balanced against the inherent risks of continued intuba- tion of support historically used to wean the patient from the ven- tion, and a clear plan should be developed outlining how post- tilator gradually.12 Multiple indices have been devised for deter- poning extubation will alter the situation for the better. mining a patient’s readiness for unassisted breathing, with variable The patient with traumatic brain injury (TBI) presents a degrees of success. Perhaps the optimal method might involve an unique problem, for two reasons. First, the patient is often unable
• 10.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 10 to maintain an adequate airway. Second, the patient is often Table 2—Causes of Failed Spontaneous unable to clear upper and lower airway secretions adequately and Breathing Trials thus is at risk for continued aspiration and pneumonia. In patients whose mental status is altered but who are expected to recover quickly, the risk imposed by 1 or 2 additional days of intubation Cause of Failure Treatment is relatively small, and delaying extubation to wait for mental sta- tus to clear can be justified. In TBI patients whose recovery is Anxiety/agitation Judicious use of benzodiazepines ± anticipated to take months (if it happens at all), the options are haloperidol early tracheostomy [see Tracheostomy, below] and attempted extu- Infection (pulmonary or Diagnosis and treatment of bation. Ultimately, the choice between these two options should extrapulmonary) causative infection be made on a case-by-case basis. Although tracheostomy is often Electrolyte abnormalities (low K+, Correction of electrolyte preferred, these patients can frequently be extubated without sig- low PO4- ) concentrations nificant sequelae.17 Pulmonary edema/cardiac ischemia Administration of diuretics ± nitrates Failed SBT Aggressive nutritional support (via When a patient fails an SBT, the first priority is to determine Deconditioning/malnutrition enteral route whenever possible) Physical therapy the reason for the failure. Initially, it is important to distinguish between patients who fail to meet the extubation criteria for unre- Aggressive bronchopulmonary lated reasons (e.g., agitation or cerebral storming in a TBI patient) hygiene Neuromuscular disease (critical ill- Specific treatment of myasthenia and those who truly are not ready to be liberated from the venti- ness polyneuropathy, myasthenia gravis (pyridostigmine, steroids, lator. Direct observation of the patient during the SBT, if feasible, gravis) plasmapheresis) may provide insight into the failed attempt. The next step is to Early consideration of tracheostomy attempt to identify the specific cause of the failure. In these situa- Increased intra-abdominal pressure Semirecumbent positioning tions, a thorough knowledge of the patient’s history is important, (obesity, abdominal distention) Nasogastric decompression with special attention paid to age, comorbid conditions, reasons for and duration of mechanical ventilation, other indices of criti- Hypothyroidism Thyroid replacement cal illness, and nutritional status [see Table 2]. Failure is frequently Large hydrothoraces (rarely primary multifactorial, and actions to improve one or more of these fac- cause but may increase work of Initiation of diuresis ± thoracentesis tors should be undertaken to alter the outcome of the next SBT. breathing in patients with marginal reserve) After a failed attempt, it is best to provide a stable, nonfatiguing form of respiratory support (e.g., PSV) until the following day, Bronchodilator therapy Excessive auto-PEEP (COPD, thus allowing the patient a period of rest. I.V. sedation to prevent agitation asthma) and air trapping Patients who persistently fail SBTs are typically classified as exhibiting failure to wean. Comorbid conditions, including con- Reduction of carbohydrate intake gestive heart failure, chronic lung disease, and renal or hepatic Excessive minute ventilation Treatment of underlying cause (e.g., insufficiency, should be treated medically to the extent possible requirements: lactate production or renal failure) ↑ CO2 production Consideration of HCO3- replace- before further trials are attempted. The excess sodium and water ment if wasting is present (e.g., frequently administered to critically ill patients can have negative Metabolic acidosis with renal tubular acidosis or pan- effects on pulmonary mechanics, making liberation from the ven- creatic fistula) tilator more difficult. Hydrostatic pulmonary edema, chest wall or visceral edema, and pleural effusions can have a greater impact on patients recovering from critical illness, who may be malnourished weeks or longer, may benefit from a more gradual reduction in and deconditioned. ventilatory support. In these cases, the term weaning is probably General measures to facilitate weaning include judicious use of an accurate description of the process. A planned gradual reduc- diuretics, upright positioning, correction of electrolyte abnormal- tion in support may be better tolerated with scheduled decreas- ities (in particular, low serum potassium or phosphate levels), es in pressure support or, alternatively, with gradual increases in and, in some cases, drainage of hydrothoraces. In addition, atten- the duration of periods of unassisted breathing. In the setting of tion should be given to providing appropriate nutritional support chronic ventilator dependence, neither method is necessarily while avoiding excessive carbohydrate administration (which can superior to the other.What is important is that the patient should increase CO2 production). Physical therapy should begin as soon never be allowed to continue until exhaustion during the wean- as possible to prevent further muscle atrophy. In addition to these ing process. Formulating a well-defined plan for weaning is more general measures, consideration should be given to performing a important than choosing a particular weaning method. One tracheostomy. Although there is disagreement about the optimal approach is to use a daily workout calendar similar to those used timing and method of tracheostomy in this setting (see below), it for athletic training, so that the details of the plan are clear and is obvious that this measure can greatly facilitate attempts to dis- easily understandable by the patient, the family, the nursing continue mechanical ventilation by eliminating the risks associat- staff, the respiratory staff, and the ICU team. Finally, both ed with extubation. When the tracheostomy is in place and all patients and families should be counseled to expect weaning to general measures have been considered, SBTs should resume. last a long time. A very small percentage of patients are unable to tolerate the PROTOCOL-BASED VENTILATOR MANAGEMENT sudden reduction in support that occurs when they are treated with continuous positive airway pressure (CPAP) or a tra- Once respiratory failure has resolved, the patient’s readiness for cheostomy collar. These patients, who are often severely decon- extubation is assessed. Traditionally, this assessment has been ditioned after having been mechanically ventilated for several made by the physician. Unfortunately, both physician factors and
• 11.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 11 TRACHEOSTOMY patient factors can result in unnecessary delays that prolong the duration of mechanical ventilation. As a result, many institutions The decision as to when a patient can be liberated from the have developed respiratory therapist–driven and nurse-driven mechanical ventilator is often complicated by concerns about the protocols that allow SBTs to occur without physician input. risks associated with failure and reintubation. Prolonged endotra- Implementation of these protocols has been shown to shorten cheal intubation can injure the airway and result in airway edema, ventilator time, reduce the incidence of ventilator-associated making reintubation difficult. In addition, the possibility that rein- pneumonia, and lower costs.18-20 tubation may be required at night or at other times when qualified Although respiratory therapist–driven or nurse-driven proto- personnel may not be readily available contributes to delays in cols improve outcome when applied to a population of ventilat- extubation. The main theoretical advantage of tracheostomy in ed patients, individual patient characteristics should still be a this context is that the issue of airway stability can be separated consideration when deciding on extubation in a given case. from the issue of readiness for extubation, and this separation may Successful completion of an SBT simply implies that the patient hasten the physician’s decision to discontinue mechanical ventila- has the ventilatory capacity to breathe spontaneously; it does not tion.Tracheostomy has other potential advantages as well, includ- guarantee that the airway is stable or that the patient can ade- ing decreased work of breathing, avoidance of continued vocal quately clear tracheobronchial secretions. The decision to end cord injury, improved bronchopulmonary hygiene, patient com- ventilatory support must rest on a careful assessment of the risks fort, and improved patient communication.32-34 On the other of continued intubation and mechanical ventilation against those hand, it has several disadvantages, including the long-term risk of of failed extubation. tracheal stenosis and the significant procedure-related complica- tion rate (reported to be between 4% and 36%).35-37 Despite the AIRWAY ASSESSMENT potential risks, it is generally believed that in properly selected Once a patient has successfully completed an SBT, a second patients, tracheostomy placement may assist in liberation from the assessment should be made to determine the need for continued mechanical ventilator, though the evidence supporting this belief airway support with the endotracheal tube. In patients who were is incosistent.38-41 initially intubated for a condition necessitating airway protection A relatively new technique that is rapidly gaining acceptance is (e.g., altered level of consciousness or angioedema), the condition the use of a percutaneous dilational approach to tracheostomy should be resolved before extubation. Alternatively, patients who placement at the bedside. Several commercially available kits are have been ventilated for long periods may have secondary airway available, and the long-term complication rate is similar to that edema related to fluid resuscitation or vocal cord damage sec- seen with open tracheostomy.36,37,42 Percutaneous access is ob- ondary to the use of the translaryngeal tube.21 tained with a needle, followed by serial dilation of the tracheoto- One way of determining whether the patient will have signif- my over a guide wire. Many advocate doing this procedure under icant airway compromise is to perform the so-called cuff leak bronchoscopic guidance, which may be associated with a lower test. In this test, the patient is placed on VCV, and the volume of periprocedural complication rate.43 The main advantage of the air lost during a single respiratory cycle when the cuff is deflat- percutaneous dilational approach is that it avoids the delays asso- ed is measured. Although the cuff leak test was originally ciated with obtaining OR time and obviates the risks associated described as a qualitative test in children with croup, it has since with patient transport.44 proved to have some value as a quantitative test, affording a The main controversy surrounding tracheostomy in this set- degree of improvement in the ability to predict which patients ting is whether there is any benefit to performing the procedure will have significant postextubation upper airway edema.22-24 early in the course of critical illness.To answer this question, sev- Leak values below 9% to 15% of the inspired volume have been eral authors compared early tracheostomy (generally < 1 week) associated with increased rates of postextubation stridor and the to prolonged translaryngeal intubation or, alternatively, to late need for reintubation.24-26 Unfortunately, the cuff leak test is an tracheostomy (> 1 week).38,40,41,45-47 Outcome measures includ- imperfect predictor. In one study, a cutoff value of 15% had a ed duration of mechanical ventilation, number of episodes of positive predictive value of only 25% for reintubation.24 With ventilator-associated pneumonia, length of ICU stay, and mor- this cutoff value, a patient would have a 75% chance of success- tality. Unfortunately, many of these reports were limited by their ful extubation in the absence of an air leak. Results such as these observational design, their small sample size, or their inclusion suggest that the cuff leak test is best used as a tool for assessing of only certain specific patient subgroups (e.g., medical or neu- risk and that failure on this test should not be an absolute con- rosurgical patients). Furthermore, surgeons have traditionally traindication to extubation. Patients who fail the test may be held very strong biases with regard to the use and timing of tra- candidates for postextubation adjunctive measures, such as cheostomy, and these biases tend to impede the conduct of ran- racemic epinephrine (to enlarge the upper airway aperture, heli- domized, controlled trials. One group carried out a multicenter um-oxygen mixtures (to decrease airflow resistance), or both. randomized trial evaluating the use of tracheostomy in trauma Although evidence supporting the use of such measures in and nontrauma patients who were expected to need mechanical adults is lacking, there are some data to support their use in chil- ventilation for more than 7 days. In this trial, early tracheostomy dren to treat upper airway obstruction and postextubation stri- offered no overall benefit; however, these results may be of lim- dor.27-30 Parenterally administered steroids have been employed ited applicability, in that the authors reported experiencing sig- to prevent postextubation stridor in children, with some success, nificant difficulty in obtaining the participation of physicians but there is no evidence that they reduce the rate of reintubation and institutions, mostly because of strong physician bias.40 As a in adults.31 Finally, patients who require prolonged intubation result, no clear consensus has yet been reached regarding the because of airway instability should be considered for tra- optimal use, timing, and method of tracheostomy in the ICU cheostomy (see below). setting.
• 12.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 12 Discussion Special Problems in Ventilator Management The expiratory flow limitation observed in patients with COPD can result in incomplete expiration, air trapping, and intrinsic pos- ACUTE LUNG INJURY AND ACUTE RESPIRATORY DISTRESS itive end-expiratory pressure (PEEPi, also referred to as auto- SYNDROME PEEP).These results can further impair respiration by decreasing ALI and ARDS are common problems in the ICU and carry a the effective contribution of diaphragmatic contraction. Although high mortality [see 8:5 Pulmonary Insufficiency].48 Over the past reducing PEEPi can be difficult, there are several maneuvers that decade, an improved understanding of the pathophysiology of should be attempted initially to decrease air trapping and the ventilator-induced lung injury has resulted in significant changes resulting PEEPi. The first is to maximize the expiratory time by in ventilator management in patients with ARDS. It is now clear increasing the inspiratory flow rate or decreasing the respiratory that the volume and pressure associated with mechanical ventila- rate. A reduction in tidal volume is also effective, in that less time tion can induce and perpetuate lung injury and the systemic is required for complete expiration of a smaller volume. Another inflammatory response syndrome (SIRS). As a result, the primary maneuver is to administer bronchodilators liberally so as to mini- goals of ventilator management in ALI and ARDS patients are mize airflow limitation. If this measure is ineffective, adjusting the (1) to avoid repetitive expansion and collapse of recruitable alveo- ventilator’s PEEP setting to match the PEEPi should be consid- lar units and (2) to avoid overdistention of functioning alveoli ered for patients who are using a spontaneous mode of ventilation. (volutrauma). This will minimize the increased work necessary to trigger the ven- Several prospective trials have employed a low-VT ventilation tilator [see Figure 5]. Extubation can generally be accomplished strategy to prevent ventilator-induced lung injury.49-53 The largest once the acute episode resolves; it should be kept in mind that of these trials compared a lung-protective approach that used a VT blood gas parameters were unlikely to have been normal at base- of 6 ml/kg of predicted body weight with a more traditional line. It is also reasonable to attempt a short period of bilevel posi- approach that used a VT of 12 ml/kg. Mortality was reduced by tive-pressure ventilation immediately after extubation in patients 22% in the patients treated with lower tidal volumes, and both with severe COPD in an effort to avoid having to reintubate these SIRS and the alveolar inflammatory response were attenuated. patients. Accordingly, a low-VT, lung-protective strategy is now standard for BRONCHOPLEURAL FISTULAS treatment of patients with ALI and ARDS.49 Bronchopleural (or parenchymal pleural) fistulas (BPFs) are an High versus Low PEEP in ARDS infrequent but severe complication of thoracic trauma or pul- ARDS is characterized by a heterogeneously distributed loss of monary resection.The incidence ranges from 2% to 12%, and the functioning alveoli, with normally compliant, open alveoli mixing condition is associated with a high mortality.58-60 Ventilator man- with collapsed, nonrecruited alveoli.54 In the setting of inadequate agement of BPF patients is difficult because large air leaks through PEEP, a lung-protective ventilation strategy may contribute to fur- the fistula, which may represent the path of least resistance to air- ther alveolar collapse and may perpetuate lung injury secondary to flow, limit adequate alveolar ventilation. Care is complicated by the repetitive opening and closing of alveolar units.55 It follows that need for a higher Paw to maintain oxygenation, which may only higher levels of PEEP may prevent the injury associated with this increase flow through the fistula. phenomenon and perhaps improve patient outcome. Additionally, The determinants of flow through a BPF have important impli- a frequent point of controversy in managing the hypoxemic patient cations for ventilator management and have been investigated in is whether it is preferable to employ higher levels of PEEP with animal models and in several small case series.61-64 Transpul- lower levels of FIO2 or lower levels of PEEP with higher levels of monary pressure and fistula resistance are the main factors influ- FIO2. Many intensivists prefer to use higher levels of PEEP to keep encing the size of the air leak. Specifically, Paw appears to have the FIO2 below 0.6, believing that higher oxygen levels may induce greatest impact on flow through the fistula; there is little or no hyperoxic lung injury. However, a randomized, controlled trial that change when peak inspiratory pressure is varied.62,65 Intrapleural compared a high-PEEP strategy with a high-FIO2 strategy found pressure plays a role as well. Conventional chest tube management the outcomes to be similar.56 Thus, it is not currently possible to involves the use of –10 or –20 cm H2O suction to evacuate the determine the optimal PEEP for ALI or ARDS patients with any pleural space and promote pleural apposition. In the setting of a certainty; the values chosen should be based on individual patient significant BPF, this measure increases the pressure gradient and response. may increase the size of the air leak.62,64 There is no clearly beneficial method of ventilator management CHRONIC OBSTRUCTIVE PULMONARY DISEASE in BPF patients, but most would agree that limitation of Paw COPD is an uncommon indication for the initiation of mechan- should be the first step and that early extubation is indicated when- ical ventilation in surgical patients; however, COPD may be pres- ever possible. Various techniques for managing respiratory failure ent in varying degrees as a comorbid condition in ventilated associated with large BPFs have been described in single case patients and may thereby complicate ventilator management. reports or small case series.65-75 High-frequency jet ventilation Ventilated patients with COPD may be oxygen or steroid depen- (HFJV) to support patients with BPF has been advocated on the dent and typically are in a tenuous state even in the absence of a basis of the low Paw generated with this ventilatory mode. Several surgical insult.The basic principles of management are (1) to treat case reports have described a decrease in fistula airflow with sub- the underlying cause of respiratory failure (e.g., sepsis or trauma), sequent closure of the BPF after the institution of HFJV.66,69,74-76 (2) to minimize airway hyperreactivity through generous use of Unfortunately, although this technique may be useful in cases of bronchodilators and steroids when necessary, (3) to manage tra- isolated BPF, most of the reports do not involve patients with cheobronchial secretions aggressively, and (4) to minimize the simultaneous lung parenchymal disease.The elevated Paw required work of breathing.57 to treat the profound hypoxemia associated with ALI and ARDS
• 13.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 13 a 20 Pressure (cm H2O) 15 Pressure 10 Trigger –2 cm H2O 5 3 b 20 Total Pressure (cm H2O) 15 Pressure PEEPi PEEP Trigger Trigger 3 cm 8 cm –5 cm 10 –2 cm H2O H2O H2O H2O 8 5 3 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (sec) Figure 5 Illustrated is intrinsic PEEP during assisted spontaneous ventilation. In a, the patient is breathing spontaneously on VCV in the assist-control mode. The respiratory rate is 20 breaths/min, and the effective I/E ratio is 1:2 (inspiration, 1 sec; expiration, 2 sec). Expiratory time is adequate; therefore, no intrinsic PEEP (PEEPi) develops, and total PEEP is equal to the set PEEP (5 cm H2O). In b, a patient who is breathing spontaneously on identical ventilator settings requires a longer expiratory phase to prevent air trapping and auto-PEEP Incomplete expiration results in an auto-PEEP equivalent to 3 cm H2O . and a new total PEEP of 8 cm H2O. This results in increased Paw, but more important, the patient must now generate 5 cm H2O of negative pressure during inspiration to trigger the ventilator. may negate any potential benefit associated with HFJV.61,63 noted (see above), PEEP prevents derecruitment. This not only Besides HFJV, a number of other unconventional modes of improves oxygenation but also has the theoretical advantage of ventilation have been used to manage BPF patients, including the preventing the repetitive collapse and reinflation of lung units use of a double-lumen tube with a variable flow resistor,71 airway associated with ventilator-induced lung injury.To date, there have pressure-release ventilation (APRV),72 high-frequency conven- been no randomized trials assessing outcome after PC-IRV, but tional ventilation,70 and differential lung ventilation using any the available data suggest that this mode is unlikely to improve combination of these modalities.66,70,75,77 Intermittent occlusion of patient outcomes in the setting of adequate PEEP. the chest tube during inspiration in an effort to lower transpul- PRONE VENTILATION monary pressure has also been reported to assist ventilation in cer- tain cases.65 Bronchial blockade78 and other newer therapies (e.g., The finding of dependent atelectasis in patients with ARDS bronchial stenting79,80 and endobronchial application of fibrin or makes prone positioning an attractive therapeutic option.54 Advo- other tissue sealants81,82) may be useful in refractory cases. Ideally, cates cite several theoretical benefits, including recruitment of pre- large BPFs should undergo early operative revision if the injury viously collapsed alveoli, relief of diaphragmatic pressure secondary and the patient’s condition permit. to the abdominal viscera, and improved drainage of tracheo- bronchial secretions.88 Several prospective randomized trials have evaluated the use of prone positioning in adults and children with Alternative Modes of Ventilation and Adjunctive Therapies ARDS.89-91 Despite consistent improvements in oxygenation, no significant benefit in terms of mortality has yet been demonstrated. INVERSE-RATIO VENTILATION In practice, prone positioning is performed intermittently Inverse-ratio pressure control ventilation (PC-IRV) has been throughout the day for periods ranging from 4 to 12 hours, with used to treat patients with ARDS and severe hypoxemia for sever- the remaining time spent in the supine position. The primary risk al decades.83,84 This mode makes use of a prolonged inspiratory associated with prone positioning is accidental removal of the time to deliver a pressure-limited breath that results in an invert- endotracheal tube, a chest tube, or an intravenous line. It is possi- ed I/E ratio (e.g., 2:1, 3:1, or 4:1 rather than the normal 1:3).The ble to place patients with an open abdomen prone in some cases; rationale for PC-IRV is based on two principles. The first princi- however, this remains technically challenging. Newer patient beds ple is that a prolonged inspiratory time results in better gas distri- designed specifically for prone positioning (e.g., Rotoprone; KCI, bution with a lower peak inspiratory pressure; the second is that San Antonio, Texas) facilitate prone ventilation, but they remain the elevated Paw improves alveolar recruitment, resulting in im- costly and are not universally available. proved oxygenation. The theoretical advantages notwithstanding, Multiply injured patients with increased intracranial pressure both animal models of ALI and several observation studies in can pose a difficult challenge. Hypoxemia is associated with a humans that compared normal-ratio ventilation with PC-IRV worse neurologic outcome in this population, but because of the have failed to find any significant advantages to the latter when increased intracranial pressure that may occur, prone positioning patients are ventilated with appropriate levels of PEEP.85-87 As is contraindicated. In this situation, limited axial rotation (i.e.,
• 14.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 14 continuous partial axial rotation to varying degrees, depending on variations in regional alveolar time constants.101 Elevation of trans- patient tolerance) is an option that may yield similar improve- pulmonary pressure above the upper inflection point is avoided ments in oxygenation.92 because the sustained Paw is generally below the peak or plateau Other injuries (e.g., unstable spinal fractures or pelvic or long- pressures observed in conventional PCV or VCV. Compared with bone fractures that call for traction) can be significant problems modes that induce a high Paw (e.g., HFOV and PC-IRV), APRV and are relative contraindications to prone positioning. In patients has the advantage of allowing spontaneous respiration, thereby with ARDS and profound hypoxemia, they are generally sec- eliminating the need for neuromuscular blocking agents. ondary concerns and should not prevent the use of prone posi- APRV has been studied extensively in animal models of tioning as an adjunctive therapy. ALI,101-104 but human data are limited to case series and a few observational studies that reported primarily on application and HIGH-FREQUENCY OSCILLATORY VENTILATION safety.105-109 In practice, three variables are manipulated: the high For several decades, high-frequency oscillatory ventilation continuous pressure (Phigh), which ranges from 20 to 30 cm H2O; (HFOV) has been used to treat respiratory failure in premature the release period, which generally lasts no longer than 1 second; infants.93 More recently, it has been used in adults with ARDS, and the low release pressure (Plow), which is between 0 and 5 cm both as primary treatment and as rescue therapy for severe hypox- H2O.The Phigh and Plow settings can be manipulated along with the emia.94-96 HFOV differs from HFJV in that it employs an oscillat- FIO2 to achieve adequate arterial oxygenation [see Figure 6]. The ing piston or diaphragm that provides high-frequency, low-ampli- release phase can be changed with respect to both duration and tude ventilation superimposed on an elevated Paw. Its primary the- frequency to ensure adequate ventilation. Spontaneous respira- oretical rationale is based on the so-called open lung approach to tions occur and are superimposed on the high CPAP. limiting ventilator-induced lung injury: it avoids the repetitive alve- No significant complications have been associated with APRV. olar opening and closing that can occur at low airway pressures There is a theoretical potential for hemodynamic compromise; while also avoiding the overdistention that occurs at higher airway however, because of the high levels of PEEP and the high plateau pressures. pressures already observed in patients with severe ARDS, Paw is Paw is determined by the operator and is maintained by contin- generally lower with APRV, and as a result, cardiovascular function uous gas flow from the ventilator through a resistance valve at the may actually improve.108 The theoretical benefits of APRV make it end of the circuit. The oscillating piston or diaphragm lies per- an attractive alternative to conventional ventilation. Un- pendicular to the gas flow. The oscillating frequency is set by the fortunately, this mode of ventilation has not been directly com- clinician and generally ranges from 3 to 6 Hz (180 to 360 breaths/ pared with low–tidal volume, lung-protective ventilation. APRV min). As in conventional ventilation, oxygenation is achieved by probably will not be widely used until it is shown to have a bene- adjusting Paw, as well as FIO2. Ventilation is altered by changing ficial effect on outcome. both the frequency and the amplitude of the piston’s oscillation. NONINVASIVE POSITIVE-PRESSURE VENTILATION HFOV differs from conventional ventilation in that increases in ventilation are achieved by reducing the frequency of oscillation Noninvasive positive-pressure ventilation (NPPV) is positive- and increasing the amplitude, which allow more time for the expi- pressure ventilation administered through either a nasal or a full- ratory phase of piston displacement. It also differs in that the expi- face mask in the form of either CPAP or bilevel positive airway ratory phase is assisted by the backward piston movement. pressure (BiPAP). PEEP is generally set at 5 to 10 cm H2O The primary side effect of HFOV is potential hemodynamic (CPAP), with or without additional pressure support at levels compromise secondary to the elevated Paw. Pneumothorax has also ranging from 5 to 20 cm H2O (BiPAP), and titrated to keep the been associated with HFOV, but it may be a marker of the severity respiratory rate under 25 breaths/min. In comparison with med- of lung injury rather than a consequence of the mode of ventila- ical therapy alone, NPPV has been demonstrated to reduce both tion.94-96 In addition, most patients require neuromuscular blocking the need for intubation and mortality in patients with exacerbated agents to prevent spontaneous respiratory effort while being main- COPD.57 Patients in whom medical therapy fails—as signaled by tained on the oscillating ventilator.The requirement for a specialized worsening tachypnea, hypoxemia, and respiratory acidosis—are adult oscillating ventilator and the general lack of familiarity with candidates for NPPV. This mode should be used in conjunction HFOV has hindered broad application of this technique. with other measures, such as inhaled bronchodilators, inhaled Despite these limitations, several studies have shown HFOV to steroids, and oral or parenteral steroids and antibiotics when be well tolerated by patients with severe ARDS. In addition, con- appropriate.57 sistent improvements in oxygenation have been documented when NPPV has sometimes been used to treat respiratory failure asso- patients are converted from conventional ventilation to HFOV.95-98 ciated with cardiogenic pulmonary edema.110-113 In this setting, As a result, HFOV is a potential option for patients with profound NPPV has certain theoretical benefits, including decreased work of hypoxemia, either alone (as rescue therapy) or in combination with breathing, decreased left ventricular afterload,114 and, possibly, other adjunctive measures (e.g., nitric oxide [NO] inhalation and decreased preload as a result of the positive intrathoracic pressure. prone ventilation). Improved oxygenation, however, is not neces- Nevertheless, the available data are inconclusive. One randomized sarily correlated with improved mortality.The role of HFOV in pri- trial was stopped early because a higher rate of myocardial isch- mary treatment of early ARDS remains to be determined.99 emia was recorded in the BiPAP group than in the CPAP group.110 Another study reported no significant cardiac events in either AIRWAY PRESSURE-RELEASE VENTILATION group.111 Both modes of NPPV, however, were superior to standard APRV is a relatively new mode of ventilation that, like HFOV, is oxygen therapy in preventing the need for intubation.111 These based on the open lung approach to limiting ventilator-induced studies are limited by their sample sizes, and as a result, it is diffi- lung injury. APRV employs a high CPAP, which is intermittently cult to draw definitive conclusions. It is possible that NPPV may released for short periods to allow lung emptying.100 The primary have a beneficial effect on outcome in patients with cardiogenic rationale is based on the idea that optimal alveolar recruitment pulmonary edema, but whether CPAP or BiPAP is therapeutically depends both on differential alveolar opening pressures and on optimal has yet to be determined.
• 15.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 15 Peak Inspiratory Gas Flow 100 Inspiratory 80 60 Spontaneous Release 40 Breaths Phase Gas Flow (L/min) Begins 20 0 -20 25% 40% Expiratory -40 50% -60 75% -80 100% Peak Expiratory Gas Flow -100 Thigh Tlow Figure 6 Illustrated is airway pressure- Phigh release ventilation, a ventilation mode 30 that maintains a high CPAP with inter- mittent release periods. The operator sets 25 the high and low pressure settings (Phigh, Airway Pressure Paw Plow) and the release time (Tlow). The 20 (cm H2O) Spontaneous mode has several theoretical advantages 15 Breaths over conventional ventilation, including 10 lower alveolar distending pressures (in patients with ARDS), avoidance of alveo- 5 lar derecruitment, and the ability for the Plow patient to breathe spontaneously. 0 1 2 3 4 5 6 7 8 9 Time (sec) NITRIC OXIDE INHALATION The use of NPPV to treat hypoxemic respiratory failure that is not due to cardiogenic pulmonary edema (e.g., respiratory failure NO is a potent vasodilator that is administered as an inhaled resulting from pneumonia, thoracic trauma, ALI, or ARDS) has gas. The delivery mode allows NO to exert its vasodilatory effect also been addressed in several small randomized trials and a sys- on pulmonary arterioles supplying ventilated (and only ventilated) tematic review.115 The results of the individual studies vary some- ˙ ˙ alveoli, thereby improving V/ Q matching, reducing intrapulmon- what, but the overall indication is that there may be some benefit ary shunting, and increasing oxygenation.118,119 Because of its to be derived from reducing the need for mechanical ventilation, short half-life in the circulation (100 milliseconds), inhaled NO is the duration of the ICU stay, and mortality in certain patient pop- not associated with systemic vasodilation and hypotension.120 ulations—particularly those in whom intubation portends a par- NO is administered through a specialized delivery system in ticularly poor prognosis (e.g., immunocompromised patients and doses of 5 to 40 parts per million. Several randomized trials have recent lung transplant recipients).115 shown it to yield modest improvements in oxygenation in adults There are several relative contraindications to NPPV, including and children with ARDS.121-124 There are also numerous reports of severe facial deformity or trauma (either of which can prevent NO administration in conjunction with prone positioning,125 sealing around the edges of the mask), a decreased level of con- HFOV,126,127 and HFJV,76 as well as in patients with BPFs.76 Un- sciousness that is not expected to improve with improved ventila- fortunately, the use of NO has not been demonstrated to improve tion, hemodynamic instability, a need for endotracheal intubation mortality. On the other hand, there have been no reports of signif- for other reasons (e.g., airway protection or surgery), and a recent icant adverse effects.The high cost of NO administration has kept upper GI operation. In addition, NPPV should not be performed it from being routinely used in ARDS patients. by physicians, therapists, or nurses unfamiliar with its use or in SURFACTANT REPLACEMENT situations where patient monitoring is inadequate. Inappropriate performance of NPPV may lead to unrecognized patient intoler- Surfactant production in adult patients with ARDS is both di- ance, worsening respiratory status, and delays in endotracheal minished and abnormal.128,129 The loss of surfactant may con- intubation.116 tribute to the alveolar collapse, intrapulmonary shunting, and Complications of NPPV include skin breakdown over the mask hypoxemia seen in ARDS. Theoretically, surfactant replacement pressure areas, gastric distention, ventilator asynchrony, and would prevent derecruitment and improve lung compliance and treatment failure (associated with worsening mental status, aspi- oxygenation. In addition, prevention of alveolar collapse is impor- ration, and delayed endotracheal intubation). In addition, NPPV tant in limiting ventilator-induced lung injury. As a result, surfac- should generally be avoided as a method of preventing reintuba- tant replacement is an attractive therapeutic option for patients tion in the ICU, because it has been associated with a worse out- with ARDS. come.117 Despite these limitations, NPPV is generally well toler- Several methods of surfactant replacement in adults have been ated and should be considered as a treatment option in certain assessed in randomized, multicenter placebo-controlled trials. In patient populations with acute respiratory failure. an early trial of aerosolized surfactant given to patients with sep-
• 16.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 16 sis-induced ARDS, no benefit was observed with respect to either approximately 50%.134,135 Vascular access may be obtained via oxygenation indices or mortality.130 This trial was limited both by either a venoarterial or a venovenous circuit. The latter is typical- the particular surfactant formulation used and by the failure to ly used in adult patients; the former is preferred in patients with employ a low–tidal volume, lung-protective ventilation strategy. A marginal cardiovascular function. subsequent report, summarizing data from two large trials of a In some centers, ECLS is employed when the risk of death with protein C–based surfactant formulation given for 24 hours, doc- continuing conventional ventilation is more than 90% and the pri- umented modest improvements in oxygenation during surfactant mary process is reversible. In 2004, one group reported on their administration. Neither trial, however, demonstrated any signifi- use of ECLS to treat 255 patients with severe ARDS (defined as a cant impact on mortality.131 In view of the lack of mortality ben- P/F ratio lower than 100 when FIO2 was 1.0, an alveolar-arterial efit, it is probably best to limit the use of surfactant in adults with gradient higher than 600 mm Hg, or a transpulmonary shunt frac- ARDS to the setting of clinical trials. tion greater than 30% despite maximal ventilatory support). More than half (53%) of the patients survived to hospital discharge.136 EXTRACORPOREAL LIFE SUPPORT Although these results are difficult to interpret in the absence of a Extracorporeal life support (ECLS) is the use of a modified control group, the mortality appears to be lower than would be heart-lung machine to support gas exchange while allowing the expected with traditional ventilator management. diseased lung to rest. The theoretical advantage of this technique The main contraindications to the use of ECLS are advanced is that it should avoid the oxygen toxicity and the volutrauma or age, malignancy, severe neurologic injury, and mechanical ventila- barotrauma that may accompany mechanical ventilation in pa- tion lasting longer than 5 to 7 days. Complications include bleed- tients experiencing severe respiratory failure. ECLS has become ing (heparinization is required), hemolysis, cerebral infarction, the standard treatment of severe respiratory failure in neonates132 renal failure, infection, and venous thrombosis. In view of the risks, and has also been used in the pediatric population.133 In adult ECLS should be reserved for patients at specialized centers who patients with severe respiratory failure, it has been employed have profound hypoxemia that is refractory to other, less invasive mainly in specialized centers, with reported survival rates of measures. References 1. Esteban A, Anzueto A, Alia I, et al: How is Collaborative Group. N Engl J Med 332:345, 24. De Bast Y, De Backer D, Moraine JJ, et al:The cuff mechanical ventilation employed in the intensive 1995 leak test to predict failure of tracheal extubation care unit? An international utilization review. Am J 14. Esteban A, Alia I,Tobin MJ, et al: Effect of sponta- for laryngeal edema. Intensive Care Med 28:1267, Respir Crit Care Med 161:1450, 2000 neous breathing trial duration on outcome of at- 2002 2. Linderholm H: Lung mechanics in sitting and hor- tempts to discontinue mechanical ventilation. Span- 25. Jaber S, Chanques G, Matecki S, et al: Post-extu- izontal postures studied by body plethysmograph- ish Lung Failure Collaborative Group. Am J Respir bation stridor in intensive care unit patients: risk ic methods. Am J Physiol 204:85, 1963 Crit Care Med 159:512, 1999 factors evaluation and importance of the cuff leak 3. Navajas D, Farre R, Rotger MM, et al: Effect of 15. Meade M, Guyatt G, Cook D, et al: Predicting test. Intensive Care Med 29:69, 2003 body posture on respiratory impedance. J Appl success in weaning from mechanical ventilation. 26. Sandhu RS, Pasquale MD, Miller K, et al: Mea- Physiol 64:194, 1988 Chest 120(6 suppl):400S, 2001 surement of endotracheal tube cuff leak to predict 4. Mutoh T, Lamm WJ, Embree LJ, et al: Abdominal 16. Girou E: Prevention of nosocomial infections in postextubation stridor and need for reintubation. J distension alters regional pleural pressures and acute respiratory failure patients. Eur Respir J Am Coll Surg 190:682, 2000 chest wall mechanics in pigs in vivo. J Appl Physiol Suppl 42:72s, 2003 27. Kemper KJ, Ritz RH, Benson MS, et al: Helium- 70:2611, 1991 oxygen mixture in the treatment of postextubation 17. Coplin WM, Pierson DJ, Cooley KD, et al: Impli- 5. Suter PM, Fairley B, Isenberg MD: Optimum cations of extubation delay in brain-injured patients stridor in pediatric trauma patients. Crit Care Med end-expiratory airway pressure in patients with meeting standard weaning criteria. Am J Respir 19:356, 1991 acute pulmonary failure. N Engl J Med 292:284, Crit Care Med 161:1530, 2000 28. Rodeberg DA, Easter AJ, Washam MA, et al: Use 1975 of a helium-oxygen mixture in the treatment of post 18. Ely EW, Bennett PA, Bowton DL, et al: Large 6. Dreyfuss D, Saumon G:Ventilator-induced lung in- scale implementation of a respiratory therapist-dri- extubation stridor in pediatric patients with burns. jury: lessons from experimental studies. Am J ven protocol for ventilator weaning. Am J Respir J Burn Care Rehabil 16:476, 1995 Respir Crit Care Med 157:294, 1998 Crit Care Med 159:439, 1999 29. Weber JE, Chudnofsky CR, Younger JG, et al: A 7. Gattinoni L, D’Andrea L, Pelosi P, et al: Regional 19. Dries DJ, McGonigal MD, Malian MS, et al: randomized comparison of helium-oxygen mix- effects and mechanism of positive end-expiratory Protocol-driven ventilator weaning reduces use of ture (Heliox) and racemic epinephrine for the pressure in early adult respiratory distress syn- mechanical ventilation, rate of early reintubation, treatment of moderate to severe croup. Pediatrics drome. JAMA 269:2122, 1993 and ventilator-associated pneumonia. J Trauma 107:E96, 2001 8. Bruhn A, Hernandez G, Bugedo G, et al: Effects 56:943, 2004 30. Hess D, Chatmongkolchart S:Techniques to avoid of positive end-expiratory pressure on gastric intubation: noninvasive positive pressure ventila- 20. Marelich GP, Murin S, Battistella F, et al: Protocol mucosal perfusion in acute respiratory distress tion and Heliox therapy. Int Anesthesiol Clin 38: weaning of mechanical ventilation in medical and syndrome. Crit Care 8:R306, 2004 161, 2000 surgical patients by respiratory care practitioners 9. Sassoon CS, Foster GT: Patient-ventilator asyn- and nurses: effect on weaning time and incidence 31. Meade MO, Guyatt GH, Cook DJ, et al: Trials of chrony. Curr Opin Crit Care 7:28, 2001 of ventilator-associated pneumonia. Chest corticosteroids to prevent postextubation airway 10. Murray MJ, Cowen J, DeBlock H, et al: Clinical 118:459, 2000 complications. Chest 120(6 suppl):464S, 2001 practice guidelines for sustained neuromuscular 21. Darmon JY, Rauss A, Dreyfuss D, et al: Evaluation 32. Astrachan DI, Kirchner JC, Goodwin WJ Jr: blockade in the adult critically ill patient. Crit Care of risk factors for laryngeal edema after tracheal Prolonged intubation vs. tracheotomy: complica- Med 30:142, 2002 extubation in adults and its prevention by dexa- tions, practical and psychological considerations. 11. Hamwi GJ: Therapy: changing dietary concepts. methasone: a placebo-controlled, double-blind, Laryngoscope 98:1165, 1988 Diabetes Mellitus: Diagnosis and Treatment. Ameri- multicenter study. Anesthesiology 77:245, 1992 33. Diehl JL, El Atrous S, Touchard D, et al: Changes can Diabetes Association, New York, 1964 22. Adderley RJ, Mullins GC: When to extubate the in the work of breathing induced by tracheotomy 12. Hall JB, Wood LD: Liberation of the patient from croup patient: the “leak” test. Can J Anaesth 34: in ventilator-dependent patients. Am J Respir Crit mechanical ventilation. JAMA 257:1621, 1987 304, 1987 Care Med 159:383, 1999 13. Esteban A, Frutos F, Tobin MJ, et al: A compari- 23. Miller RL, Cole RP: Association between reduced 34. Davis K Jr, Campbell RS, Johannigman JA, et al: son of four methods of weaning patients from cuff leak volume and postextubation stridor. Chest Changes in respiratory mechanics after tracheosto- mechanical ventilation. Spanish Lung Failure 110:1035, 1996 my. Arch Surg 134:59, 1999
• 17.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 17 35. Freeman BD, Isabella K, Lin N, et al: A meta- 54. Gattinoni L, Caironi P, Pelosi P, et al: What has 75. Wippermann CF, Schranz D, Baum V, et al: analysis of prospective trials comparing percuta- computed tomography taught us about the acute Independent right lung high frequency and left neous and surgical tracheostomy in critically ill respiratory distress syndrome? Am J Respir Crit lung conventional ventilation in the management patients. Chest 118:1412, 2000 Care Med 164:1701, 2001 of severe air leak during ARDS. Paediatr Anaesth 36. Melloni G, Muttini S, Gallioli G, et al: Surgical 55. Richard JC, Maggiore SM, Jonson B, et al: Influ- 5:189, 1995 tracheostomy versus percutaneous dilatational tra- ence of tidal volume on alveolar recruitment: 76. Campbell D, Steinmann M, Porayko L: Nitric cheostomy: a prospective-randomized study with respective role of PEEP and a recruitment maneu- oxide and high frequency jet ventilation in a patient long-term follow-up. J Cardiovasc Surg (Torino) ver. Am J Respir Crit Care Med 163:1609, 2001 with bilateral bronchopleural fistulae and ARDS. 43:113, 2002 56. Brower RG, Lanken PN, MacIntyre N, et al: Can J Anaesth 47:53, 2000 37. Kearney PA, Griffen MM, Ochoa JB, et al: A sin- Higher versus lower positive end-expiratory pres- 77. Crimi G, Candiani A, Conti G, et al: Clinical gle-center 8-year experience with percutaneous sures in patients with the acute respiratory distress applications of independent lung ventilation with dilational tracheostomy. Ann Surg 231:701, 2000 syndrome. N Engl J Med 351:327, 2004 unilateral high-frequency jet ventilation (ILV- 38. Boynton JH, Hawkins K, Eastridge BJ, et al:Trache- 57. Celli BR, MacNee W: Standards for the diagnosis UHFJV). Intensive Care Med 12:90, 1986 ostomy timing and the duration of weaning in and treatment of patients with COPD: a summary 78. Otruba Z, Oxorn D: Lobar bronchial blockade in patients with acute respiratory failure. Crit Care of the ATS/ERS position paper. Eur Respir J 23: bronchopleural fistula. Can J Anaesth 39:176, 8:R261, 2004 932, 2004 1992 39. Maziak DE, Meade MO, Todd TR: The timing of 58. Wright CD,Wain JC, Mathisen DJ, et al: Postpneu- 79. Watanabe S, Shimokawa S, Yotsumoto G, et al: tracheotomy: a systematic review. Chest 114:605, monectomy bronchopleural fistula after sutured The use of a Dumon stent for the treatment of a 1998 bronchial closure: incidence, risk factors, and bronchopleural fistula. Ann Thorac Surg 72:276, 40. Sugerman HJ,Wolfe L, Pasquale MD, et al: Multi- management. J Thorac Cardiovasc Surg 112:1367, 2001 center, randomized, prospective trial of early tra- 1996 80. Tayama K, Eriguchi N, Futamata Y, et al: cheostomy. J Trauma 43:741, 1997 59. Sonobe M, Nakagawa M, Ichinose M, et al: Analy- Modified Dumon stent for the treatment of a 41. Rumbak MJ, Newton M, Truncale T, et al: A sis of risk factors in bronchopleural fistula after bronchopleural fistula after pneumonectomy. Ann prospective, randomized, study comparing early pulmonary resection for primary lung cancer. Eur Thorac Surg 75:290, 2003 percutaneous dilational tracheotomy to prolonged J Cardiothorac Surg 18:519, 2000 81. Lin J, Iannettoni MD: Closure of bronchopleural translaryngeal intubation (delayed tracheotomy) 60. de Perrot M, Licker M, Robert J, et al: Incidence, fistulas using albumin-glutaraldehyde tissue adhe- in critically ill medical patients. Crit Care Med 32: risk factors and management of bronchopleural sive. Ann Thorac Surg 77:326, 2004 1689, 2004 fistulae after pneumonectomy. Scand Cardiovasc J 82. Glover W, Chavis TV, Daniel TM, et al: Fibrin glue 42. Freeman BD, Isabella K, Cobb JP, et al: A 33:171, 1999 application through the flexible fiberoptic bron- prospective, randomized study comparing percu- 61. Bishop MJ, Benson MS, Sato P, et al: Comparison choscope: closure of bronchopleural fistulas. J taneous with surgical tracheostomy in critically ill of high-frequency jet ventilation with conventional Thorac Cardiovasc Surg 93:470, 1987 patients. Crit Care Med 29:926, 2001 mechanical ventilation for bronchopleural fistula. 83. Tharratt RS, Allen RP, Albertson TE: Pressure 43. Dulguerov P, Gysin C, Perneger TV, et al: Percu- Anesth Analg 66:833, 1987 controlled inverse ratio ventilation in severe adult taneous or surgical tracheostomy: a meta-analysis. 62. Walsh MC, Carlo WA: Determinants of gas flow respiratory failure. Chest 94:755, 1988 Crit Care Med 27:1617, 1999 through a bronchopleural fistula. J Appl Physiol 84. Wang SH,Wei TS:The outcome of early pressure- 44. Waydhas C: Intrahospital transport of critically ill 67:1591, 1989 controlled inverse ratio ventilation on patients with patients. Crit Care 3:R83, 1999 63. Spinale FG, Linker RW, Crawford FA, et al: Con- severe acute respiratory distress syndrome in sur- 45. Arabi Y, Haddad S, Shirawi N, et al: Early trache- ventional versus high frequency jet ventilation with gical intensive care unit. Am J Surg 183:151, 2002 ostomy in intensive care trauma patients improves a bronchopleural fistula. J Surg Res 46:147, 1989 85. Neumann P, Berglund JE, Andersson LG, et al: resource utilization: a cohort study and literature 64. Tilles RB, Don HF: Complications of high pleur- Effects of inverse ratio ventilation and positive review. Crit Care 8:R347, 2004 al suction in bronchopleural fistulas. Anesthesiol- end-expiratory pressure in oleic acid-induced lung 46. Bouderka MA, Fakhir B, Bouaggad A, et al: Early ogy 43:486, 1975 injury. Am J Respir Crit Care Med 161:1537, tracheostomy versus prolonged endotracheal intu- 2000 65. Powner DJ, Grenvik A:Ventilatory management of bation in severe head injury. J Trauma 57:251, life-threatening bronchopleural fistulae: a summa- 86. Ludwigs U, Philip A: Pulmonary epithelial perme- 2004 ry. Crit Care Med 9:54, 1981 ability and gas exchange: a comparison of inverse 47. Brook AD, Sherman G, Malen J, et al: Early versus ratio ventilation and conventional mechanical ven- 66. Mortimer AJ, Laurie PS, Garrett H, et al: late tracheostomy in patients who require pro- tilation in oleic acid-induced lung injury in rabbits. Unilateral high frequency jet ventilation: reduction longed mechanical ventilation. Am J Crit Care Chest 113:459, 1998 of leak in bronchopleural fistula. Intensive Care 9:352, 2000 Med 10:39, 1984 87. Huang CC, Shih MJ, Tsai YH, et al: Effects of 48. Milberg JA, Davis DR, Steinberg KP, et al: inverse ratio ventilation versus positive end-expira- 67. Carlon GC, Ray C Jr, Klain M, et al: High-fre- Improved survival of patients with acute respirato- tory pressure on gas exchange and gastric intra- quency positive-pressure ventilation in manage- ry distress syndrome (ARDS): 1983–1993. JAMA mucosal PCO(2) and pH under constant mean ment of a patient with bronchopleural fistula. 273:306, 1995 airway pressure in acute respiratory distress syn- Anesthesiology 52:160, 1980 drome. Anesthesiology 95:1182, 2001 49. Ventilation with lower tidal volumes as compared 68. Litmanovitch M, Joynt GM, Cooper PJ, et al: with traditional tidal volumes for acute lung injury 88. Pelosi P, Brazzi L, Gattinoni L: Prone position in Persistent bronchopleural fistula in a patient with and the acute respiratory distress syndrome. The acute respiratory distress syndrome. Eur Respir J adult respiratory distress syndrome: treatment 20:1017, 2002 Acute Respiratory Distress Syndrome Network. N with pressure-controlled ventilation. Chest Engl J Med 342:1301, 2000 104:1901, 1993 89. Kornecki A, Frndova H, Coates AL, et al: A ran- 50. Amato MB, Barbas CS, Medeiros DM, et al: domized trial of prolonged prone positioning in 69. McGuire GP: Lung ventilation and bronchopleur- children with acute respiratory failure. Chest Effect of a protective-ventilation strategy on mor- al fistula. Can J Anaesth 43:1275, 1996 tality in the acute respiratory distress syndrome. N 119:211, 2001 70. Feeley TW, Keating D, Nishimura T: Independent 90. Gattinoni L, Tognoni G, Pesenti A, et al: Effect of Engl J Med 338:347, 1998 lung ventilation using high-frequency ventilation prone positioning on the survival of patients with 51. Brochard L, Roudot-Thoraval F, Roupie E, et al: in the management of a bronchopleural fistula. Tidal volume reduction for prevention of ventila- acute respiratory failure. N Engl J Med 345:568, Anesthesiology 69:420, 1988 2001 tor-induced lung injury in acute respiratory dis- tress syndrome. The Multicenter Trail Group on 71. Carvalho P, Thompson WH, Riggs R, et al: Man- 91. Guerin C, Gaillard S, Lemasson S, et al: Effects of Tidal Volume Reduction in ARDS. Am J Respir agement of bronchopleural fistula with a variable- systematic prone positioning in hypoxemic acute Crit Care Med 158:1831, 1998 resistance valve and a single ventilator. Chest respiratory failure: a randomized controlled trial. 111:1452, 1997 JAMA 292:2379, 2004 52. Brower RG, Shanholtz CB, Fessler HE, et al: Pros- pective, randomized, controlled clinical trial com- 72. Darwish RS, Gilbert TB, Fahy BG: Management 92. Staudinger T, Kofler J, Mullner M, et al: Compari- paring traditional versus reduced tidal volume of a bronchopleural fistula using differential lung son of prone positioning and continuous rotation ventilation in acute respiratory distress syndrome airway pressure release ventilation. J Cardiothorac of patients with adult respiratory distress syn- patients. Crit Care Med 27:1492, 1999 Vasc Anesth 17:744, 2003 drome: results of a pilot study. Crit Care Med 53. Stewart TE, Meade MO, Cook DJ, et al: 73. Charan NB, Carvalho CG, Hawk P, et al: Indepen- 29:51, 2001 Evaluation of a ventilation strategy to prevent dent lung ventilation with a single ventilator using 93. Bhuta T, Henderson-Smart DJ: Elective high-fre- barotrauma in patients at high risk for acute respi- a variable resistance valve. Chest 107:256, 1995 quency oscillatory ventilation versus conventional ratory distress syndrome. Pressure- and Volume- 74. Turnbull AD, Carlon G, Howland WS, et al: High- ventilation in preterm infants with pulmonary dys- Limited Ventilation Strategy Group. N Engl J Med frequency jet ventilation in major airway or pul- function: systematic review and meta-analyses. 338:355, 1998 monary disruption. Ann Thorac Surg 32:468, 1981 Pediatrics 100:E6, 1997
• 18.
  © 2005 WebMD,Inc. All rights reserved. ACS Surgery: Principles and Practice 8 CRITICAL CARE 6 MECHANICAL VENTILATION — 18 94. Mehta S, Granton J, MacDonald RJ, et al: High- lease ventilation on renal perfusion and function in Acute and sustained effects of early administration frequency oscillatory ventilation in adults: the patients with acute lung injury. Intensive Care Med of inhaled nitric oxide to children with acute respi- Toronto experience. Chest 126:518, 2004 28:1426, 2002 ratory distress syndrome. Pediatr Crit Care Med 95. Mehta S, Lapinsky SE, Hallett DC, et al: Prospec- 110. Mehta S, Jay GD,Woolard RH, et al: Randomized, 5:469, 2004 tive trial of high-frequency oscillation in adults with prospective trial of bilevel versus continuous posi- 125. Johannigman JA, Davis K Jr, Miller SL, et al: acute respiratory distress syndrome. Crit Care Med tive airway pressure in acute pulmonary edema. Prone positioning and inhaled nitric oxide: syner- 29:1360, 2001 Crit Care Med 25:620, 1997 gistic therapies for acute respiratory distress syn- 96. Derdak S, Mehta S, Stewart TE, et al: High-fre- 111. Park M, Sangean MC,Volpe MdeS, et al: Random- drome. J Trauma 50:589, 2001 quency oscillatory ventilation for acute respiratory ized, prospective trial of oxygen, continuous posi- 126. Varkul MD, Stewart TE, Lapinsky SE, et al: distress syndrome in adults: a randomized, con- tive airway pressure, and bilevel positive airway Successful use of combined high-frequency oscilla- trolled trial. Am J Respir Crit Care Med 166:801, pressure by face mask in acute cardiogenic pul- tory ventilation, inhaled nitric oxide, and prone 2002 monary edema. Crit Care Med 32:2407, 2004 positioning in the acute respiratory distress syn- 97. Cartotto R, Ellis S, Gomez M, et al: High frequen- 112. Delclaux C, L’Her E, Alberti C, et al:Treatment of drome. Anesthesiology 95:797, 2001 cy oscillatory ventilation in burn patients with the acute hypoxemic nonhypercapnic respiratory 127. Mehta S, MacDonald R, Hallett DC, et al: Acute acute respiratory distress syndrome. Burns 30:453, insufficiency with continuous positive airway pres- oxygenation response to inhaled nitric oxide when 2004 sure delivered by a face mask: a randomized con- combined with high-frequency oscillatory ventila- 98. Fort P, Farmer C, Westerman J, et al: High-fre- trolled trial. JAMA 284:2352, 2000 tion in adults with acute respiratory distress syn- quency oscillatory ventilation for adult respiratory 113. Masip J, Betbese AJ, Paez J, et al: Non-invasive drome. Crit Care Med 31:383, 2003 distress syndrome—a pilot study. Crit Care Med pressure support ventilation versus conventional 128. Hallman M, Spragg R, Harrell JH, et al: Evidence 25:937, 1997 oxygen therapy in acute cardiogenic pulmonary of lung surfactant abnormality in respiratory fail- 99. Wunsch H, Mapstone J: High-frequency ventila- oedema: a randomised trial. Lancet 356:2126, ure: study of bronchoalveolar lavage phospho- tion versus conventional ventilation for treatment 2000 lipids, surface activity, phospholipase activity, and of acute lung injury and acute respiratory distress 114. Lenique F, Habis M, Lofaso F, et al: Ventilatory plasma myoinositol. J Clin Invest 70:673, 1982 syndrome. Cochrane Database Syst Rev (1): and hemodynamic effects of continuous positive 129. Gunther A, Ruppert C, Schmidt R, et al: Surfact- CD004085, 2004 airway pressure in left heart failure. Am J Respir ant alteration and replacement in acute respiratory 100. Garner W, Downs JB, Stock MC, et al: Airway Crit Care Med 155:500, 1997 distress syndrome. Respir Res 2:353, 2001 pressure release ventilation (APRV): a human trial. 115. Keenan SP, Sinuff T, Cook DJ, et al: Does nonin- 130. Anzueto A, Baughman RP, Guntupalli KK, et al: Chest 94:779, 1988 vasive positive pressure ventilation improve out- Aerosolized surfactant in adults with sepsis- 101. Stock MC, Downs JB, Frolicher DA: Airway pres- come in acute hypoxemic respiratory failure? A induced acute respiratory distress syndrome. sure release ventilation. Crit Care Med 15:462, systematic review. Crit Care Med 32:2516, 2004 Exosurf Acute Respiratory Distress Syndrome 1987 116. Wood KA, Lewis L, Von Harz B, et al: The use of Sepsis Study Group. N Engl J Med 334:1417, noninvasive positive pressure ventilation in the em- 1996 102. Neumann P, Hedenstierna G: Ventilatory support by continuous positive airway pressure breathing ergency department: results of a randomized clini- 131. Spragg RG, Lewis JF, Walmrath HD, et al: Effect improves gas exchange as compared with partial cal trial. Chest 113:1339, 1998 of recombinant surfactant protein C-based surfac- ventilatory support with airway pressure release 117. Esteban A, Frutos-Vivar F, Ferguson ND, et al: tant on the acute respiratory distress syndrome. N ventilation. Anesth Analg 92:950, 2001 Noninvasive positive-pressure ventilation for respi- Engl J Med 351:884, 2004 103. Martin LD, Wetzel RC, Bilenki AL: Airway pres- ratory failure after extubation. N Engl J Med 350: 132. Stolar CJ, Snedecor SM, Bartlett RH: Extra- sure release ventilation in a neonatal lamb model of 2452, 2004 corporeal membrane oxygenation and neonatal acute lung injury. Crit Care Med 19:373, 1991 118. Rossaint R, Falke KJ, Lopez F, et al: Inhaled nitric respiratory failure: experience from the extracor- oxide for the adult respiratory distress syndrome. poreal life support organization. J Pediatr Surg 26: 104. Smith RA, Smith DB; Does airway pressure 563, 1991 release ventilation alter lung function after acute N Engl J Med 328:399, 1993 lung injury? Chest 107:805, 1995 119. Dellinger RP, Zimmerman JL, Taylor RW, et al: 133. Swaniker F, Kolla S, Moler F, et al: Extracorporeal Effects of inhaled nitric oxide in patients with acute life support outcome for 128 pediatric patients 105. Rasanen J, Cane RD, Downs JB, et al: Airway pres- with respiratory failure. J Pediatr Surg 35:197, sure release ventilation during acute lung injury: a respiratory distress syndrome: results of a random- ized phase II trial. Inhaled Nitric Oxide in ARDS 2000 prospective multicenter trial. Crit Care Med 19:1234, 1991 Study Group. Crit Care Med 26:15, 1998 134. Kolla S, Awad SS, Rich PB, et al: Extracorporeal 120. Moncada S, Palmer RM, Higgs EA: Nitric oxide: life support for 100 adult patients with severe res- 106. Sydow M, Burchardi H, Ephraim E, et al: Long- piratory failure. Ann Surg 226:544, 1997 term effects of two different ventilatory modes on physiology, pathophysiology, and pharmacology. oxygenation in acute lung injury: comparison of Pharmacol Rev 43:109, 1991 135. Michaels AJ, Schriener RJ, Kolla S, et al: Extracor- airway pressure release ventilation and volume- 121. Taylor RW, Zimmerman JL, Dellinger RP, et al: poreal life support in pulmonary failure after trau- controlled inverse ratio ventilation. Am J Respir Low-dose inhaled nitric oxide in patients with ma. J Trauma 46:638, 1999 Crit Care Med 149:1550, 1994 acute lung injury: a randomized controlled trial. 136. Hemmila MR, Rowe SA, Boules TN, et al: Extra- 107. Davis K Jr, Johnson DJ, Branson RD, et al: Airway JAMA 291:1603, 2004 corporeal life support for severe acute respiratory pressure release ventilation. Arch Surg 128:1348, 122. Krafft P, Fridrich P, Fitzgerald RD, et al: Effective- distress syndrome in adults. Ann Surg 240:595, 1993 ness of nitric oxide inhalation in septic ARDS. 2004 108. Kaplan LJ, Bailey H, Formosa V: Airway pressure Chest 109:486, 1996 release ventilation increases cardiac performance 123. Sokol J, Jacobs SE, Bohn D: Inhaled nitric oxide in patients with acute lung injury/adult respiratory for acute hypoxic respiratory failure in children distress syndrome. Crit Care 5:221, 2001 and adults: a meta-analysis. Anesth Analg 97:989, Acknowledgment 109. Hering R, Peters D, Zinserling J, et al: Effects of 2003 spontaneous breathing during airway pressure re- 124. Fioretto JR, de Moraes MA, Bonatto RC, et al: Figures 1, 2, 4, 5, and 6 Seward Hung.