Weaning patients from mechanical ventilation remains one of the most challenging aspects of intensive ...... respiratory technician; PT patient;. PaO2 arterial ...
CHAPTER 47 Victor Kim and Gerard J. Criner
Weaning from Mechanical Ventilation CHAPTER OUTLINE
LEARNING OBJECTIVES
Learning Objectives Case Study Determining the Cause of Respiratory Failure When is the Patient Ready to Wean? Predictors of Weaning Outcome
After studying this chapter, you should be able to:
Pulmonary Gas Exchange Respiratory Muscle Strength P0.1 Work of Breathing Gastric Tonometry Mixed Venous Oxygen Content Respiratory Pattern during Spontaneous Breathing
Specific Weaning Methods Trials of Spontaneous Breathing Intermittent Mandatory Ventilation Pressure-Support Ventilation Efficacy of Different Weaning Techniques
■■ Determine when a patient is ready to begin the weaning process, based on clinical history, physical examination, and routine laboratory data. ■■ Use bedside weaning parameters to predict weaning outcome. ■■ Postulate a differential diagnosis of common and uncommon causes of weaning failure. ■■ Understand the advantages and disadvantages of the various weaning techniques. ■■ Realize that certain techniques, such as noninvasive ventilation after extubation and daily interruption of sedatives, can increase your likelihood of liberating the patient from mechanical ventilation.
Techniques to Aid Weaning Tracheostomy Daily Interruption of Sedatives Noninvasive Ventilation Protocol or Computer-Based Strategies Adjunctive Therapy
Summary Review Questions Answers References Additional Reading
902
During the past 25 years, there has been a significant increase in the number of patients who receive mechanical ventilation as a means of life support during surgery or life-threatening medical illness. Although mechanical ventilation has clear-cut benefits, it is also associated with a significant number of complications, such as decreased cardiac output, increased intracranial pressure, ventilator-associated pneumonia (VAP), and ventilator-induced lung injury (VILI). In addition, mechanical ventilation is expensive and hinders efficient patient movement through the intensive care unit.
C HAPTE R 47 • Weaning f rom M echanical Ventilation
903
CAS E STU DY A 70-year-old nursing home resident was intubated for aspiration pneumonia and respiratory distress. He was treated with antibiotics and clinically improved over the period of 5 days. During that time, he required 100% FiO2 and needed sedatives to maintain patient comfort and patient ventilator synchrony, and frequent endotracheal tube suctioning for purulent secretions. He is now awake and alert with normal oxygen saturation on
40% FiO2 and PEEP of 5 cm H2O. Minute ventilation (MV) is 10 L/ min. He appears cachectic and has lower extremity contractures. He has no fever and is hemodynamically stable off all vasopressor support. When the patient is placed on CPAP of 5 cm H2O and pressure support (PS) of 0 cm H2O, the patient’s respiratory rate increases to 40 breaths/min and tidal volume (VT) drops to 250 mL within 1 min.
Weaning patients from mechanical ventilation remains one of the most challenging aspects of intensive care. Despite the advent of new and promising weaning indices, the skills to determine which patients should be weaned and when patients are ready for weaning remain a mix of art and science. These skills appear to be greatly enhanced by experience. About 20–25% of ventilated patients fail an initial attempt at discontinuing mechanical ventilation and will require more concentrated and prolonged attempts for discontinuance (i.e., weaning). For patients requiring prolonged mechanical ventilation, about 40% of the time spent on the ventilator is devoted to the weaning process. This percentage is even higher in patients with specific diseases such as chronic obstructive pulmonary disease (COPD), who may be undergoing active weaning attempts for as much as 60% of their total time spent receiving mechanical ventilation. In this chapter, we review the clinical parameters that determine which patients are ready to begin the weaning process, the interpretation of bedside parameters used to predict weaning outcome, and the merits and disadvantages of specific weaning techniques to successfully discontinue mechanical ventilation.
DETERMINING THE CAUSE OF RESPIRATORY FAILURE Before mechanical ventilation can be safely withdrawn, the abnormality causing respiratory failure must be identified and show favorable signs of responding to treatment. To identify the physiologic causes of respiratory failure, it is useful to separate the causes into three major categories: (1) hypoxemic respiratory failure, (2) ventilatory pump failure, and (3) psychologic factors ( 47-1). Hypoxemic respiratory failure can result from hypoventilation shunting, impaired pulmonary gas exchange, or decreased mixed venous blood oxygen content. The chest radiograph, physical examination, and alveolar-arterial oxygen gradient are useful in distinguishing among intrapulmonary shunting, increased physiologic deadspace, and alveolar hypoventilation as possible causes of hypoxemic respiratory failure (see Chap. 15). Ventilatory pump dysfunction is considered by some authors to be the most common cause of failure to wean from mechanical ventilation. Failure of the respiratory system to sustain adequate ventilation to meet the demands imposed by the illness may occur whenever respiratory demand exceeds ventilatory pump capacity. Ventilatory pump failure may occur because of an increased ventilatory load (even in patients with normal respiratory system), resulting from increased deadspace, hypermetabolism due to sepsis and/or fever, or increased CO2 production due to increased carbohydrate load. In contrast, normal or only slightly elevated respiratory loads may not be sustained by subjects with decreased ventilatory pump capacity due to impaired central respiratory drive, phrenic nerve dysfunction, thoracic wall abnormalities, or severe derangements of respiratory muscle function (e.g., underlying neuromuscular disease, electrolyte disturbances). Abnormalities of central respiratory drive can be seen in patients with structural injury to the central nervous system, overuse of sedative agents, and metabolic alkalosis. Diaphragm
Before withdrawing mechanical ventilation, the cause of respiratory failure must be identified.
Respiratory pump dysfunction is considered the most common cause of failure to wean.
904
V. K im and G.J. C riner
TABLE 47-1 CAUSES OF RESPIRATORY FAILURE
Hyperinflation is a frequently overlooked cause of weaning failure.
Hypoxemic respiratory failure Impaired pulmonary gas exchange (shunt, V/Q mismatch) Pneumonia Congestive heart failure Pulmonary embolism ARDS Decreased mixed venous oxygen content Congestive heart failure Ventilatory pump failure Decreased minute ventilation with a relatively normal respiratory workload Thoracic wall abnormalities (flail chest, rib fractures) Peripheral neurologic disorder (phrenic nerve damage, critical care illness polyneuropathy) Muscular dysfunction (neuromuscular blocking agent-associated weakness/myopathy) Central respiratory depression (drug overdose, brainstem infarction) Severe metabolic alkalosis Increased respiratory workload with poor ability to maintain adequate ventilation Increased minute ventilation requirements (sepsis, fever, hyperthermia, increased CO2 production) Increased deadspace Increased elastic workload (decreased lung and/or chest wall compliance, dynamic hyperinflation) Increased resistive workload (airway obstruction, secretions, endotracheal tube, ventilator circuit) Psychologic factors Anxiety Depression
dysfunction can be seen in patients with cold-induced phrenic nerve injury or direct diaphragm injury that may occur during cardiothoracic surgery. Diaphragm dysfunction has also been reported in patients following upper abdominal surgery. Impaired respiratory muscle function can also result from various underlying medical conditions. Hyperinflation occurs in patients with acute exacerbations of severe asthma or COPD and is frequently overlooked as a cause of weaning failure. It causes a shortening in the diaphragm’s precontraction length, which causes the diaphragm to work on a disadvantageous portion of its tension–length curve. Hyperinflation also alters the orientation of the diaphragmatic fibers medially inward, and decreases the length of the zone of apposition, factors that further decrease the diaphragm’s force-generating capacity. Other disorders commonly encountered in the intensive care unit may cause abnormal respiratory muscle function, thereby hindering weaning. These include undernutrition, electrolyte disturbances (hypophosphatemia, hypokalemia, hypocalcemia, hypomagnesemia), and thyroid dysfunction. Recently, diaphragm muscle atrophy has been shown to occur as a consequence of inactivity in previously healthy brain-dead organ donors receiving fully assisted mechanical ventilation for periods of only 24–96 h.1
WHEN IS THE PATIENT READY TO WEAN? Resolution or significant improvement in the underlying cause of respiratory failure is the most important prerequisite before weaning is attempted.
Before an attempt is made to wean a patient, certain prerequisites should be met (Table 47-2). The most important prerequisite appears to be resolution or significant improvement in the underlying cause of respiratory failure. Patients should be hemodynamically stable, with minimal or no need for vasopressor agents. The absence of sepsis or hyperthermia should be confirmed. Sedative drugs and neuromuscular blocking agents should be discontinued. Patients should be awake, alert, and able to manage secretions and protect their airway. Significant fluid, electrolyte, and metabolic disorders should be corrected before weaning attempts are made. Adequate gas exchange marked by a PaO2 to FiO2 ratio greater than 200, FiO2 requirements of 50% or less, and positive end-expiratory pressure 5 cm H2O or less are desirable. Adequate respiratory muscle strength needs to be ensured (maximum inspiratory pressure [MIP] or negative inspiratory force at least −25 cm H2O).
C HAPTE R 47 • Weaning f rom M echanical Ventilation
Resolution or significant improvement of the underlying cause of respiratory failure Stable hemodynamic state Absence of sepsis or hyperthermia Cessation of sedative drugs Cessation of neuromuscular blocking agents Cessation of vasopressor agents Patients should be awake, alert, and able to manage secretions and protect their airway Correction of metabolic and electrolyte disorders Adequate gas exchange PaO2 to FiO2 ratio greater than 200 FiO2 requirements less than 50% PEEP requirements equal to or less than 5 cm H2O Adequate respiratory muscle strength
905
TABLE 47-2 NECESSARY CONDITIONS TO DECIDE WHEN A PATIENT IS READY FOR WEANING
PaO2 partial pressure of arterial oxygen; FiO2 inspired fraction of oxygen; PEEP positive end-expiratory pressure
PREDICTORS OF WEANING OUTCOME Determining when a patient is ready to wean from the ventilator is a complicated task. Considerable research has been devoted to finding variables that predict weaning outcome.
Pulmonary Gas Exchange The adequacy of pulmonary gas exchange must be assessed before initiating weaning. In the past, several indices derived from arterial blood gas analysis have been used to predict weaning outcome. These indices are derived from retrospective studies and, consequently, have limitations. An arterial blood to inspired O2 ratio (PaO2/FiO2) greater than 238 has a positive predictive value of 90%, yet its negative predictive value is only 10%.2 In another study, an arterial/alveolar O2 tension of 0.35 had positive and negative predictive values slightly greater than 0.5.3 Although adequate arterial oxygenation is essential to initiate weaning, it is clear that the predictive value of this index by itself is insufficient to predict weaning outcome.
Adequate oxygenation is essential to initiate weaning, but its predictive value regarding weaning outcome is poor.
Respiratory Muscle Strength The strength and endurance of the respiratory system seem to be major determinants of weaning outcome. Sahn and Lakshminarayan were among the first to describe the use of simple bedside criteria to assist decisions in discontinuing ventilatory support.4 In a study involving 100 patients, these investigators measured resting MV, maximum voluntary ventilation (MVV) (i.e., maximum sustainable ventilation over 15 s... times 4, MVV), and MIP with a spirometer. Of these, 76 patients who had an MV less than 10 L/min, MIP of −30 cm H2O or less, and MVV twice their resting MV who were able to complete a 2-h spontaneous breathing trial via an endotracheal tube were successfully extubated; seven more patients with a MIP of −25 cm H2O or less and a mean MV of 10.2 L/min and able to successfully complete at least a 2 hours weaning trial were able to be extubated, although they were not able to double their resting MV. By contrast, 17 patients with an MIP greater than −22 cm H2O who could not complete a spontaneous breathing trial could not be extubated. Application of these criteria in subsequent studies, however, did not yield comparable results. When evaluating 47 patients on mechanical ventilation, Tahvanainen et al found that using a MIP less than −30 cm H2O as a weaning predictor produced a false-negative result in 11 of 11 patients and a false-positive result in 8 of 23 patients.5 Similarly, other authors reported poor negative and positive predictive values when evaluating other weaning parameters such as vital capacity (VC), minute ventilation (VE), and MVV. Factors that may account for the variability in bedside respiratory mechanics to predict weaning outcomes include different patient populations, variability in the duration of mechanical ventilation, different techniques used in measuring respiratory mechanics, and
Respiratory muscle strength and endurance are important determinants of weaning outcome.
906
V. K im and G.J. C riner
inability of the measurements to accurately assess the true cause of ventilatory dependency. For example, no matter what respiratory parameters are measured, if a patient develops an acute severe episode of heart failure, or bronchospasm post-extubation, respiratory failure will recur, and thus, weaning will fail. Because of the poor and variable results of bedside parameters to predict weaning outcome, investigators turned to more complicated measurements of respiratory mechanics, such as P0.1, gastric tonometry, and measurements of the work of breathing and mixed venous oxygen content.
P0.1 The airway pressure generated 100 ms after initiating an inspiratory effort against an occluded airway (P0.1) is believed to reflect central respiratory drive and has been proposed as a useful predictor of weaning outcome. The values for normal, healthy individuals are 2 cm H2O or less. Herrera et al observed that 89% of patients with a P0.1 greater than 4 cm H2O failed weaning attempts.6 In patients with COPD, Sassoon and Mahutte found that patients with a P0.1 greater than 6 cm H2O were unable to wean from ventilatory support, but patients with a P0.1 less than 6 cm H2O were successfully extubated.7 Several studies have shown a large variation in outcome when P0.1 is used, possibly because of its inability to predict endurance or its application to patient groups with different diseases causing respiratory failure. This technique also requires special equipment and trained personnel, which makes it less appealing.
Work of Breathing Work of breathing can be measured by plotting the transpulmonary pressure against VT. Transpulmonary pressure is calculated from the difference between pleural pressure (estimated with an endoesophageal balloon catheter) and the airway pressure. One study found that mechanical ventilation was necessary when work of breathing exceeded 1.8 kg/m2/min. A similar study found a discriminating value of 1.34 kg/m2/min to separate ventilator-dependent from ventilator-independent patients. At this level, the rate of false-negative and falsepositive results was 13.8%. An additional study evaluated work of breathing in a group of 17 patients, six of whom required prolonged mechanical ventilation. A work index less than 1.6 kg/m2/min was observed in all patients who had a successful weaning trial; this was more accurate than conventional weaning parameters in determining weaning outcome. Furthermore, patients who went from unsuccessful to successful weaning did not show significant improvement in conventional weaning parameters, but did show improvement in work of breathing measurements. There are no large prospective studies comparing this parameter against more conventional weaning parameters. The relative invasiveness and complexity of data acquisition and analysis, and the requirement for dedicated staff and equipment to perform the test make it unappealing as an effective clinical tool.
Gastric Tonometry It has been proposed that measurement of gastric pH during weaning can help predict weaning success. A fall in gastric pH during weaning from mechanical ventilation would indicate mucosal ischemia from hypoperfusion as a result of blood flow diverted toward the respiratory muscles to meet their increased metabolic demands. Mohsenifar et al measured gastric pH before and during weaning attempts in 29 patients who were intubated for respiratory failure.8 All patients were ventilated for more than 48 h and were treated with ranitidine. Despite similar hemodynamic parameters, changes in respiratory breathing pattern, and gas exchange during weaning, those successfully liberated from the ventilator had no change in gastric pH, whereas those that failed had a fall in gastric pH. The authors concluded that this technique can help predict weaning success. However, it is unclear if these results are applicable to other disease states or to those not treated with gastric acid suppressive therapy. Additionally, the technique requires the placement of an orogastric or nasogastric tube. These drawbacks limit the routine utility of this technique.
C HAPTE R 47 • Weaning f rom M echanical Ventilation
907
Mixed Venous Oxygen Content Mixed venous oxygen saturation (SvO2) has also been measured during attempts at weaning from mechanical ventilation. In one study, the SvO2 progressively fell during weaning in eight patients who failed a spontaneous breathing trial, compared to 11 patients who were successfully extubated and had no significant change in SvO2. The fall in SvO2 was hypothesized to result from an increased oxygen cost of breathing and increased oxygen extraction. Measurement of mixed venous oxygen content needs prospective validation prior to recommendation for routine use.
Respiratory Pattern during Spontaneous Breathing Multiple studies have reported the role of breathing pattern in the outcome of weaning from mechanical ventilation. The development of rapid shallow breathing, the presence of asynchronous or paradoxical thoracoabdominal movements, and marked accessory muscle recruitment during a spontaneous breathing trial are physical exam findings that herald an unsuccessful weaning trial. Based on the premise that patients who fail weaning trials rapidly develop a high respiratory rate and a drop in VT, Yang and Tobin combined measurements of frequency (f) and VT into the rapid shallow breathing index, f/VT.3 They obtained data in 36 patients and noticed that an f/VT ratio of 105 breaths/min/L best differentiated patients who weaned successfully from those who failed. They subsequently validated the rapid shallow breathing index in 64 patients, comparing it against conventional weaning indexes (Table 47-3). An f/VT ratio less than 105 predicted a successful weaning trial (Fig. 47-1). The positive and negative predictive values were 0.78 and 0.95, respectively. The predictive power of the f/VT ratio was better than any of its components, supporting the use of this index. The f/VT ratio is attractive because it is relatively easy to obtain and the determinant value (i.e., »100) is easy to remember. It is important to recognize that this test is most accurate in patients who have received mechanical ventilation for less than 7 days. In a subsequent study, Epstein attempted to define the cause of extubation failure in patients whose f/VT predicted weaning success.9 He analyzed 94 consecutive patients whose f/VT before the weaning trial predicted successful extubation. The f/VT was measured while patients breathed unassisted for 1 min. Of the 94 patients extubated, 84 had an f/VT less than 100 and 10 had an f/VT of 100 or more. Extubation failure, defined as the need to reintubate within 72 h, occurred in 14 of 84 patients in the group with f/VT below 100 and 4 of 10 patients with f/VT above 100 (Table 47-4). When the cause for respiratory failure was analyzed, the underlying respiratory process was responsible for extubation failure in all four patients with f/VT of 100 or more. In contrast, the initial respiratory process was the cause for extubation failure in only 1 of 14 patients with an f/VT less than 100; new problems, such as
INDEX
VALUE
Minute ventilation (L/min) Respiratory frequency (breaths/min) Tidal volume (mL) Maximal inspiratory pressure (cm H2O) Dynamic compliance (mL/cm H2O) Static compliance (mL/cm H2O) PaO2/PAO2 Frequency/tidal volume ratio (breaths/min/L) CROP index (mL/breath/min)
£15 £38 ³325
