ArticlePlus
Click on the links below to access all the ArticlePlus for this article.
Please note that ArticlePlus files may launch a viewer application outside of your web browser.
A FEW years ago, a critical care fellow embarrassed me by defending his less than stellar documentation of a patient’s physical examination with the following statement to one of my colleagues, “Dr. Hubmayr said that listening to the chest of a mechanically ventilated patient is a waste of time.” My colleague thought that I had lost my wits and that I was a terrible role model for our critical care training program. Although I felt quoted out of context and briefly contemplated taking a course in media training, I must confess to a long-held skepticism about the usefulness of lung auscultation in the management of critically ill mechanically ventilated patients. Do I listen to the chest of my patients? Absolutely! After all, one should not ignore new murmurs, extra heart sounds, or pericardial rubs. Do I listen to the lungs? Yes, but largely because patients expect me to. The stethoscope is a powerful bonding tool with patients, especially when an artificial airway prevents them from talking. Frankly, I seem to learn a lot more about breathing from palpation of neck and abdominal muscles, from inspection of chest movements, and from analysis of respiratory variations in airway and vascular pressures than I do from listening to wheezes and crackles. The report by Lichtenstein et al. in this issue of the Journal is a welcome reinforcement of my personal bias. 1
The authors compare the diagnostic performance of three techniques, namely, auscultation, bedside chest radiography, and lung ultrasonography in patients with acute respiratory distress syndrome. Thoracic computer tomography served as the definitive standard for the detection of pleural effusions, consolidation of alveolar airspaces, and alveolar-interstitial (permeability) edema. Given the choice of target conditions, it is not surprising that the stethoscope—at least in the hands of one investigator—performed poorly and had accuracy akin to the flip of a coin. The bedside frontal chest radiograph substantially underestimated the extent of lung injury, which is consistent with a large body of published work, beginning with Gattinoni et al.’ s original report on computer tomographic findings in acute respiratory distress syndrome. 2In contrast to auscultation and plain film chest radiography, lung ultrasonography had a greater than 90% diagnostic accuracy for each of the three target conditions and showed excellent spatial correlations with computer tomographic findings.
This Editorial View accompanies the following article: Lichtenstein D, Goldstein I, Mourgeon E, Cluzel P, Grenier P, Rouby J-J: Comparative diagnostic performances of auscultation, chest radiography, and lung ultrasonography in acute respiratory distress syndrome. Anesthesiology 2004; 100:9–15.
Considering these observations, is it really time to hang up the stethoscope? Not yet, but I do wish to raise a few caveats about the cost-effectiveness of lung auscultation in the intensive care setting. Much of what we know about lung sound interpretation has been handed down by generations of master clinicians, but is far from supported by a comprehensive database. There are no large-scale population studies in which the acoustic properties of the respiratory system have been assessed and validated against sensitive and specific measurements of lung structure or function. For certain, sophisticated approaches and elegant analyses of lung sound recordings in experimental animals with injured lungs suggest that correlations between acoustic energy transfer, measures of gas exchange, and lung mechanics exist. 3However, such studies merely provide a proof of concept and do not establish efficacy of pulmonary acoustics in the care of ventilator-dependent patients. Until further research establishes clinical efficacy, we should be mindful that auscultation of the lungs might drive therapeutic decisions, which could cause benefit as well as harm. Does every wheeze warrant a bronchodilator treatment, or every crackle a diuretic or an increase in positive end-expiratory pressure? Clearly not, but such decisions are often made despite the lack of reliable sensitivity and specificity data.
Ultrasound fails to penetrate the gas-containing lung tissue and therefore, to date, has found limited application in the assessment of pulmonary lesions. Nevertheless, in their report the current authors speculate that changes in the appearance of consolidated lung and of air/tissue interface artifacts, which they describe as “rockets and comet-tails,” may be useful indicators of lung recruitment. Although this hypothesis is intriguing, the efficacy of ultrasonography as a guide to positive end-expiratory pressure management will be difficult to establish. The literature on lung recruitment and its putative benefits keeps growing, but the link between surrogate physiologic endpoints oxygenation and respiratory compliance and the more meaningful outcome variables survival and lung healing has yet to be established. There is strong experimental evidence from the bench that recruitment is a desirable therapeutic goal as long as overdistension of other parts of the lung can be avoided. However, there is no agreement on how to define the optimal balance between maximal recruitment and minimal overdistension. Given its limited depth penetration into lung tissue, it is hard to imagine that ultrasonography will prove to be the gauge by which the risk/benefit balance of lung recruitment targeted interventions may be judged.
Encouraged by the superior diagnostic performance of ultrasonography in detecting acute respiratory distress syndrome-related pulmonary lesions, the current authors speculate that frequent lung examinations with handheld ultrasound probes may obviate the need for routine daily chest radiographs, bringing about considerable cost savings. They acknowledge that ultrasound is blind to malpositions of endotracheal tubes or indwelling vascular catheters, consequential findings that are detected in up to 10% of routine examinations. 4Yet, the cost-effectiveness of obtaining routine daily chest radiographs in mechanically ventilated patients remains controversial insofar as this practice has not resulted in proven reductions in hospital mortality or length of stay. 5
The focus on the lung examination is obviously important for anesthesiologists and intensivists who treat mechanically ventilated patients. I suspect, however, that the search for a more sensitive and specific lung examination will not be the only reason why several years from now my stethoscope will gather random noise while I bond with my patients via a portable ultrasound probe. I can imagine that guided by clinical pretest probability I will take a quick look at the carotids, check the patency of neck and axillary veins, take a peak at right and left ventricular ejection fractions, make sure that bile and urine are flowing without backing up in the gallbladder or renal pelvis, and work my way toward the legs looking for clots.
The trouble with this vision is that currently I am not confident that I know how to use the device in all of its proposed applications. There is a great temptation to just buy a probe and learn as one goes. I consider that to be a mistake. There are experts who can teach us intensivists how to use ultrasound. 6It will be up to us to make a commitment and learn the strengths and limitations of this technology. This is a prerequisite for testing the efficacy of portable ultrasonography as an adjunct to or even an integral part of the routine examination of critically ill patients. Maybe then I will hang up my stethoscope.
