© 2004 American Thoracic Society
Using Electric Impedance Tomography to Assess Regional Ventilation at the BedsideUniversity Hospital Uppsala, Sweden Mechanical ventilation is a cornerstone in the treatment of acute respiratory failure, but like a double-edged sword, it can both save lives and kill. It ensures adequate oxygenation if ventilation is distributed to most of the lung (or, ideally, is distributed in proportion to regional lung blood flow), but it may, and often will, cause harm to the lung itself. Unfortunately, we have no bedside technique for monitoring regional ventilation and perfusion or pending lung damage. How then can we know whether our ventilator settings are optimal? Arterial blood gas analysis and the recording of airway pressure–volume curves can guide us in the ventilator setting. Even if both techniques are available at the bedside, however, they reflect only overall lung function, and regional abnormalities may pass undetected. Regional ventilation and perfusion can be studied by isotope and magnetic resonance techniques, and aeration of the lung by computed tomography. None of these are bedside tools. In this issue of the Journal (pp. 791–800), Victorino and coworkers (1) report on measurements of regional ventilation by a truly bedside technique, electric impedance tomography, in mechanically ventilated patients suffering from acute respiratory failure. Electric impedance tomography was presented 20 years ago (2). By injecting an alternating current between sequential pairs of adjacent electrodes around the chest, voltage differences between other pairs of non-injecting electrodes can be recorded and translated into regional impedance values. Air is a poor conductor of electric current and causes high impedance, whereas water or blood is a good conductor. The difference makes it possible to detect changes in air and tissue content, enabling the assessment of ventilation distribution (3). Good agreement with ventilation measurements by fast electron-beam computed tomography (4) and ventilation scintigraphy (5) has been shown in animal experiments. For the first time, Victorino and coworkers present results in severely ill patients with marked inhomogeneity of ventilation distribution (1). There was excellent reproducibility of the measurement of ventilation distribution when partitioning the lungs into four zones, with a variation of only 4 to 7%. This is equal to, or better than, the reproducibility of most physiological measurements. Bias was minor and the difference between the impedance tomography technique and computed tomography was less than 10% for detecting imbalances between the right and left lung. This is good enough to make the technique an interesting bedside tool. The reader is encouraged to look and listen to the narrated video that can be found in the online supplement (1). The asynchronous pattern of opening and expansion of different lung regions is fascinating and educational. The video sequences suggest that the non-aerated regions are collapsed, not fluid-filled, and need a threshold airway pressure to pop open. Additional studies are required, however, to distinguish between possible mechanisms of airlessness (6). An interesting finding was that dependent lung regions did not inspire until some time had elapsed during the slow inflation, with air going initially to upper regions. This could not be explained by reopening of collapsed or flooded regions because they were already aerated. A possible explanation may be airway closure, trapping air in distal alveoli (7). This seems to be an important phenomenon that promotes formation of atelectasis during anesthesia. The higher the oxygen concentration, the faster the gas is adsorbed and atelectasis produced (8). The importance of this mechanism in acute respiratory failure remains to be determined. Another issue is why re-opening of atelectasis in the anesthetized subject with healthy lungs requires an airway pressure of 40 cm H2O (9), whereas re-opening of collapsed or flooded acutely sick lung has already started at 10–15 cm H2O (10). Regional aeration curves can be of value in studying the underlying differences between these two conditions. Can electric impedance tomography be used to avoid injurious patterns of mechanical ventilation? Cyclic opening and closing of airways and alveoli produce shear forces that can harm the lung (11). These cyclic events can be detected by electric impedance tomography; see, for example, the video sequences in the online supplement (1). Overdistension by excessive tidal volume and elevated airway pressure (volotrauma and barotrauma) has long been known to damage the lung (12). Overdistension, however, will be difficult to detect with the presently used impedance technique because the algorithms for calculating ventilation are based on changes in impedance, not on absolute values. Will the electric impedance tomography technique be improved further? Victorino and coworkers (1) tested the effect of a slight change in the electrode positioning around the chest. They used a standard set up of sixteen electrodes, although application of additional electrodes might provide more data for regional analysis (13). It would have been interesting to find out whether resolution is poorer and reconstruction artifacts larger in the core of the thorax than in the periphery, closer to the skin. This is likely. The injected current will choose the way with the lowest impedance, not necessarily the shortest distance. A catheter with an electrode positioned in the esophagus might improve central resolution, although it would be more invasive. The algorithms may also be modified (14). The transverse area of the thorax was trapezoid in the studied patients, but the algorithms are based on a circular structure. Reliable recording of absolute air content would be valuable. Electric impedance tomography may also be used for the assessment of regional lung blood flow. By using the ECG signal for gating the impedance variation, regional lung blood flow can be determined. This has been tested to distinguish left and right lung blood flow (15). A bedside tool for the adjustment of mechanical ventilation with the capacity of measuring regional ventilation and perfusion would be worth waiting for. It seems like we are close to getting it. FOOTNOTES Conflict of Interest Statement: G.H. serves as a consultant to Maquet Critical Care Inc. (formerly Siemens Life Support Systems) and received $1,000 in 2001 for speaking at a conference sponsored by Siemens. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
