Monitoring Respiratory Drive and Respiratory Muscle Unloading during Mechanical Ventilation

Transcription

1 Monitoring Respiratory Drive and Respiratory Muscle Unloading during Mechanical Ventilation J. Beck and C. Sinderby z Introduction Since Galen's description 2000 years ago that the lungs could be inflated artificially, mechanical ventilation has become a primary intervention for life support [1]. The common indication for mechanical ventilation is respiratory failure, defined as a major abnormality in gas exchange [2]. According to Esteban et al. [3], how the assist is delivered in terms of volume and respiratory rate varies widely among centers. Recent randomized clinical trials have suggested that limitation of tidal volume reduces the risk of ventilator-induced lung injury (VILI) [4]. Given that the positive pressure ventilator solely pumps air or medical gas into the lungs (negative pressure ventilation will not be discussed), one must assume that mechanical ventilation either substitutes or partially assists the respiratory muscles' function of inflating the lungs. If the patient is heavily sedated and/or paralyzed, the ventilator's delivery of positive pressure substitutes the respiratory muscles, and the settings for pressure, tidal volume and respiratory rate are decided upon by the caregiver. New evidence indicates that the practice of deep sedation has a negative impact on care and increases both the duration of hospital stay and mortality [5] (for an excellent overview see Burchardi [6]). Algorithms for sedation, therefore, increasingly favor reduced levels and daily interruption of sedation [7]. If sedation is limited, patients will likely be breathing spontaneously, involving respiratory muscle activity, and allowing tidal volumes and breathing frequency to be controlled by the patient. Intuitively, one would anticipate that the ventilator should adapt to and follow the patient's breathing pattern. This is, however, not always the case. Poor interaction between the ventilator and the patient is commonly reported and has been suggested to increase the need for sedation [8, 9]. z Respiratory Drive and Respiratory Muscle Unloading To understand the underlying issues that cause poor patient-ventilator interaction, it is important to discuss how mechanical ventilation unloads the patient and controls respiratory drive. Unloading of the respiratory muscles is an indistinct term and several factors are involved in the process; these are discussed below. The first aim during respiratory failure should be to reduce the underlying source that causes the increased load. For example, during acute bronchoconstriction, the aim would be to reduce airway resistance (reducing resistive load), which

2 Monitoring Respiratory Drive and Respiratory Muscle Unloading during Mechanical Ventilation 469 would reduce dynamic hyperinflation (improves inspiratory muscle strength), and also increase compliance (reducing elastic load). Another manner in which the respiratory muscles can be unloaded is by reducing the respiratory drive with sedation and analgesia, which will reduce respiratory muscle pressure generation. To maintain adequate ventilation during sedation, an increase in the level of ventilatory assist may be necessary to compensate for the reduced patient work. Since mechanical ventilation is applied to increase ventilation, this in itself will of course reduce respiratory drive and pressure generation if it successfully reduces CO 2 levels. If the mechanical ventilator delivers the assist when the patient's inspiratory muscles actively try to inhale, the mechanical ventilator can be considered an artificial inspiratory muscle, aiding the inspiratory muscles to generate sufficient ventilation. z How Synchronous is Mechanical Ventilation? In published clinical trials on outcome from mechanical ventilation, the differences between controlled ventilation vs partial ventilatory assist should be interpreted with caution, as it is not always stated what modes of ventilation are used or if the patient is spontaneously breathing. There have not yet been any clinical studies about the role of synchronized mechanical ventilation on patient outcome, where the assist is truly synchronized to patient effort. For example, the Cochrane review [10] on synchronized modes of mechanical ventilation does not include any quantitative index of patient ventilator interaction. In fact, the Cochrane reviewers call for evidence that modes that manufacturers refer to as `synchronous' actually do provide synchronized assist. In fact, reports on patient-ventilator interaction suggest that triggered modes of partial mechanical ventilation (e.g., pressure support ventilation) frequently are asynchronous to patient's efforts [11], especially when assist levels are high [12± 14]. Patient-ventilator asynchrony may cause the patient to `fight the ventilator' increasing both inspiratory and expiratory muscle activity (e.g., [15]), as seen in Figure 1. Since poor patient-ventilator interaction is inherent to the use of pneumatic triggering and cycling-off algorithms, intuitively, improved trigger and cycling-off could resolve issues related to patient's fighting the ventilator. As asynchrony is usually manifested by the patient making an effort to inhale when the inhalation valve is closed or the patient is exhaling when the inhalation valve is open, it is interesting to observe how newer modes like bilevel positive airway pressure (BiPAP) and airway pressure release ventilation (APRV) overcome these shortcomings by simply not occluding the patient [16, 17]. These modes simply deliver time-cycled assist switching between two pressure levels. The patient can breathe freely during both the high and low pressure level such that one part of the minute ventilation is produced by the ventilator's pressure cycling and one part is obtained by the patient's spontaneous breathing [16, 17]. Evidently, the avoidance of occlusions reduces the load on the respiratory muscles. The delivery of assist with these modes is, however, not synchronized to patient effort and it is unclear how they differ from conventional (non-triggered, time-cycled) modes in terms of unloading. Patient-ventilator asynchrony can also be manifested by the patient being passive and triggering with minimal use of the inspiratory muscles. Figure 2 shows an example of ventilator breaths that are triggered by small efforts, followed by a period of assist throughout which the diaphragm is not active. Inevitably, this pattern of

3 470 J. Beck and C. Sinderby Fig. 1. Recordings of flow, airway pressure (Paw), and transversus abdominis electromyograph (EMG) in a critically ill patient with chronic obstructive pulmonary disease (COPD) receiving pressure support (PS) of 20 cmh 2 O. The onset of expiratory muscle activity (vertical dotted line) occurred when mechanical inflation was only partly completed. From [15] with permission breathing with small negative deflections in esophageal pressure and very brief bursts of diaphragm activation may result in inactivity-induced atrophy, a field that has recently gained interest [18]. Although it is obvious from the diaphragm electrical activity tracing in Figure 2 that changing the ventilator settings (i.e., reducing the rise time, reducing the level of assist, and/or earlier cycling-off) would improve patient-ventilator synchrony ± and may increase activation ± nothing in the airway pressure tracing suggests that the patient's breathing is over assisted. To introduce further complexity on the issue of patient-ventilator asynchrony, Figure 3 shows an intubated rabbit with acute lung injury (ALI) spontaneously breathing on volume control (6 ml/kg) with no positive end-expiratory pressure (PEEP). Looking at the volume tracing there is no evidence of muscle activity. However, the diaphragm electrical activity tracing reveals that there is very high diaphragm activity during the expiration phase that is reduced during the ventilator breath. Hence, diaphragm activity is 100% asynchronous to the ventilator's assist and the high diaphragm activity indicates that the animal may well be actually `under-assisted'. Actually, in the case presented in Figure 3 addition of PEEP reduced the diaphragm activity to low levels. Generally, by only having the flow, volume and airway pressure signals available, and no feedback about the respiratory drive, it is difficult to interpret whether the ventilator's assist is appropriate.

4 Monitoring Respiratory Drive and Respiratory Muscle Unloading during Mechanical Ventilation 471 Fig. 2. Example of tracings obtained in a with chronic obstructive pulmonary disease patient, breathing on pressure support. Note that for the successfully triggered breaths, the diaphragm electrical activity (EAdi) and esophageal pressure (Pes) efforts cease immediately after the onset of the assist (Paw) and are characterized by inactivity of the diaphragm during the period of assist (grey bars). Also note the presence of wasted efforts (indicated by arrows) where the patient made a neural inspiratory effort, but failed to trigger the ventilator Fig. 3. Time tracing of diaphragm electrical activity (EAdi) and volume obtained in a non-vagotomized rabbit after HCl-induced acute lung injury in volume controlled mode. When inspiratory volume increases (grey shadowed bars) there is a suppression of the EAdi. Decreasing lung volume was related to an increase of the EAdi, producing severe asynchrony between rabbit and ventilator

5 472 J. Beck and C. Sinderby z How to Monitor Respiratory Muscle Efforts during Mechanical Ventilation How do we know that the patient is using the inspiratory muscles during mechanical ventilation? Apart from the ventilator signaling that the patient is triggering, there are no clinical methods to ensure that the inspiratory muscles are active throughout the ventilator delivered breath. It should be noted that pneumatic triggers are today so sensitive that even flow oscillations induced by the heart may auto-trigger the assist without need for inspiratory muscles [19]. The recent introduction of adjustable off-cycling is aimed at matching the termination of ventilator assist to the end of the patient's inspiratory effort; however, without feedback about neural activation, determination of the `optimal' off-cycling setting is difficult. Since there are no clinical methods available for determining the adequate physiological cycling-off criteria, it is impossible to determine when the patient's inspiration stops vis- -vis the ventilator's assist. Another enigma in the field of mechanical ventilation is the quantification of inspiratory muscle unloading. Conventionally, the only indications for successful ventilation are adequate blood gases and clinical inspection of the patient. There is no easy monitoring parameter available to quantify inspiratory muscle unloading in clinic. In other words, we know when and how much assist the ventilator delivers, and we know flow, tidal volume, and minute ventilation, however, we do not know if or how much the patient's effort was reduced by manipulating the assist. Generally, two methods are available to monitor and quantify respiratory muscle unloading: 1) diaphragm electrical activity [20] and 2) esophageal pressure [21]. Fig. 4. Example in a mechanically ventilated patient of how delivery of assist affects the neuro-mechanical coupling of the diaphragm. During an inspiratory occlusion (right tracing), the nadir of the esophageal pressure (Pes) coincides with the peak of the diaphragm activation. When a ventilator breath is delivered (left tracing), the Pes becomes `uncoupled' and the esophageal pressure nadir no longer coincides with the peak of diaphragm activation. EAdi: diaphragm electrical activity

6 Monitoring Respiratory Drive and Respiratory Muscle Unloading during Mechanical Ventilation 473 Diaphragm electrical activity represents the temporal-spatial summation of the neural respiratory drive. The esophageal pressure mirrors the pleural pressure resulting from the muscle activation. A confounding factor during mechanical ventilation, however, is how it in itself affects the transformation of neural activity into pressure, the so called neuro-mechanical coupling [22, 23]. When using esophageal pressure, the nadir of the signal is typically used to indicate the end of the inspiratory effort. As depicted in Figure 4, the nadir of the esophageal pressure coincides with the peak of the diaphragm activation during an occluded inspiration (no assist). However, during assist delivery the esophageal pressure nadir no longer coincides with the peak of diaphragm activation. Thus, the ventilator's influence on the neuro-mechanical uncoupling must be taken into consideration when using the esophageal pressure signal to infer respiratory muscle unloading. z Conclusion An important limitation with today's ventilatory management strategies is that while the timing and magnitude of assist is known, it is not possible to determine clinically the amount of unloading and reduction in respiratory drive. Moreover there are no methods to reveal if the ventilator's assist is delivered when the patient makes the inspiratory effort. To improve accuracy, new monitoring devices for determining respiratory drive and patient-ventilator synchrony are needed. Determination of respiratory drive using diaphragm electrical activity, especially in combination with esophageal pressures can readily help to determine the effect of assist delivery. References 1. Colice GL (1994) Historical perspective on the development of mechanical ventilation. In: Tobin M (ed) Principles and Practice of Mechanical Ventilation, 1 st edition, McGraw Hill, New York, pp 1±36 2. Aldrich TK, Prezant DJ (1994) Indications for mechanical ventilation. In: Tobin M (ed) Principles and Practice of Mechanical Ventilation, 1 st edition, McGraw Hill, New York, pp 155± Esteban A, Anzueto A, Alia I, et al (2000) How is mechanical ventilation employed in the intensive care unit? An international utilization review. Am J Respir Crit Care Med 161:1450± The Acute Respiratory Distress Syndrome Network (2000) Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 342:1301± Schweickert WD, Gehlbach BK, Pohlman AS, Hall JB, Kress JP (2004) Daily interruption of sedative infusions and complications of critical illness in mechanically ventilated patients. Crit Care Med 32:1272± Buchardi H (2004) Aims of sedation/analgesia. Minerva Anestesiol 70:137± Kress JP, Pohlman AS, O'Connor MF, Hall JB (2000) Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med 342:1471± Wheeler AP (1993) Sedation, analgesia, and paralysis in the intensive care unit. Chest 104:566± Burchardi H, Rathgeber J, Sydow M (1995) The concept of analgo-sedation depends on the concept of mechanical ventilation. In: Vincent JL (ed) Yearbook of Intensive Care and Emergency Medicine. Springer-Verlag, Heidelberg, pp 155±164

7 474 J. Beck and C. Sinderby: Monitoring Respiratory Drive and Respiratory Muscle Unloading 10. Greenough A, Milner AD, Dimitriou G (2001) Synchronized mechanical ventilation for respiratory support in newborn infants, Cochrane Database Syst Rev. CD Beck J, Gottfried SB, Navalesi P, et al (2001) Electrical activity of the diaphragm during pressure support ventilation in acute respiratory failure. Am J Respir Crit Care Med 164:419± Aslanian P, El Atrous S, Isabey D, et al (1998) Effects of flow triggering on breathing effort during partial ventilatory support. Am J Respir Crit Care Med. 157(1):135± Leung P, Jubran A, and Tobin MJ (1997) Comparison of assisted ventilator modes on triggering, patient effort, and dyspnea. Am J Respir Crit Care Med 155:1940± Nava S, Ambrosino N, Rubini F, et al (1993) Effect of nasal pressure support ventilation and external PEEP on diaphragmatic activity in patients with severe stable COPD. Chest. 103:143± Parthasarathy S, Jubran A, Tobin MJ (1998) Cycling of inspiratory and expiratory muscle groups with the ventilator in airflow limitation. Am J Respir Crit Care Med 158:1471± Stock MC, Downs JB (1987) Airway pressure release ventilation. Crit Care Med 15:462± Baum M, Benzer H, Putensen C, et al (1989). Biphasic positive airway pressure (BIPAP): a new form of augmented ventilation. Anaesthesist 38:452± Vassilakopoulos T, Petrof BJ (2004) Ventilator-induced diaphragmatic dysfunction. Am J Respir Crit Care Med 169:336± Hill L (2001) Flow triggering, pressure triggering, and autotriggering during mechanical ventilation. Crit Care Med 28:579± Aldrich T, Sinderby C, McKenzie D, Estenne M, Gandevia S (2002) Electrophysiologic techniques for the assessment of respiratory muscle function. ATS/ERS Statement on Respiratory Muscle Testing. Am J Respir Crit Care Med 166:518± Tobin M, Brochard L, Rossi A (2002) Assessment of Respiratory Muscle Function in the Intensive Care Unit. ATS/ERS Statement on Respiratory Muscle Testing. Am J Respir Crit Care Med 166:518± Beck J, Spahija J, Sinderby C (2003) Respiratory muscle unloading during mechanical ventilation. In: Vincent JL (ed) Yearbook of Intensive Care and Emergency Medicine. Springer, Heidelberg, pp 280± Spahija J, Beck J, Sinderby C (2004) Respiratory failure and diaphragm function. In: Vincent JL (ed) Yearbook of Intensive Care and Emergency Medicine. Springer, Heidelberg, pp 325±332