As early as 1967, it was apparent that ultrasound was ideally suited for the detection of pleural effusions.¹ In addition, thoracic ultrasonography can also detect less common pleural pathology, guide thoracentesis, and other pleural procedures. As a result, its role for critical care physicians has become increasingly important.

Ultrasound examination of the pleura is influenced by the surrounding structures. The ribs block ultrasound waves and prevent deeper structures from being visualized. In contrast, air reflects ultrasound. The surface of aerated lung will reflect most of the ultrasound waves. The point of reflection is immediately below the pleura. However, if the lung is consolidated or atelectatic, it can be readily visualized.

In addition to the artifacts seen in other aspects of medical ultrasonography, there are specific artifacts, such as rib shadowing, that are found commonly in pleural ultrasound. Air reverberation artifacts, which originate below the pleura, are another artifact type that can be commonly observed by clinicians (see Chapter 19). Translational artifacts, due to patient breathing or mechanical ventilation, may also confuse the examiner.²

Obesity and subcutaneous edema can degrade image quality. Significant edema may also present problems in judging the depth for procedures. The presence of subcutaneous air will make visualization of deeper structures problematic. The use of firm pressure on the skin and the use of a coupling medium will reduce some artifacts. Artifacts are often visible in only one scanning plane, so changing the probe angle may cause artifacts to disappear. Further, artifacts usually will not move with the respiratory cycle. The observation of an image throughout several respiratory cycles often helps to clarify the issue.

Pleural ultrasonography can be performed with many different types of two-dimensional ultrasound machines. Doppler capability is not needed. A probe with a small “footprint” to easily fit between rib spaces should be used. The preferred ultrasound probe is a phased array transducer with a frequency of 2–5 MHz (typically 3.5 MHz) that may also be used for cardiac ultrasonography. Probes with higher frequencies can visualize the pleural surface, but lack adequate penetration for clinical applications that require visualization of deeper thoracic structures.

In order to standardize image interpretation, the use of uniform probe orientation and a screen marker is required. The machine should be set up, so that the image marker on the screen is in the upper left corner of the screen. When using the longitudinal scanning plane, the probe should be oriented with the probe marker positioned cephalad. If this orientation is maintained, the cephalad direction will always be to the left of the screen.

The gain and depth should be adjusted so that the chest wall, pleural surfaces, and deeper structures, such as the liver or spleen, with overlying diaphragm are well visualized. It is recommended that the depth setting first be set to near-maximum depth, which allows for an overview of deep structures; and then can be adjusted so that the relevant target is in the center of the screen. When better visualization of the pleural surface and superficial structures is required, the depth setting can be adjusted to allow for examination of near-field structures. Alternatively, a higher frequency transducer may be used to improve resolution of near-field structures, though with resulting reduction in penetration.

The pleural ultrasound examination should be performed in a systematic fashion. With the probe applied in a rib interspace using a longitudinal scanning plane, the pleural surface appears as a bright line between the chest wall and the air artifact of the lung or the pleural effusion. By sliding the probe longitudinally along the chest wall, adjacent interspaces can be examined. After completing an entire scan line, the probe can be moved medially or laterally and another scan line can be obtained. In this way, a near-complete mapping of the pleura can be obtained.

The diaphragmatic pleura can be viewed through a trans-hepatic approach. On the patient’s left side, in the absence of pleural fluid, the full length of the diaphragmatic pleura may be difficult to identify due to the presence of aerated lung that blocks ultrasonographic visualization of the structure. The mediastinal pleura generally cannot be visualized with a transthoracic probe. The visceral subcostal pleura may be obscured by rib shadowing. Changing the probe angle or altering the patient’s position can overcome this problem.

The normal pleura is 0.2–0.4 mm thick.³ Although the frequency of the probe used for general ultrasonography does not allow for resolution of the individual parietal and visceral pleura, this does not have clinical relevance for the intensivist. A complete ultrasound examination of the pleura in the ambulatory patient is usually performed with the patient in an upright position. This poses particular problems for the intensivist, because patients are often in the supine position while mechanically ventilated and sedated. Fortunately, many pleural abnormalities can be detected via an anterior and lateral thoracic examination of a supine patient. If the posterior chest must be examined, the supine patient may be placed in a lateral decubitus position. If a major change in a patient’s position is required to perform pleural ultrasonogaphy, the intensivist must pay careful attention to support lines and tubes to avoid unplanned device removal.

Pleural effusion is a common problem in the intensive care unit (ICU). Mattison et al. reported a prevalence of 62% in medical ICU patients.⁴ The most common causes were heart failure, atelectasis, parapneumonic effusion, and hepatic hydrothorax. Malignancy accounted for 3.2% and empyema accounted for 1.6%. Compared with patients without effusions, patients with effusions are sicker and have longer ICU stays and longer durations of mechanical ventilation.

Ultrasonography is well suited for the identification and evaluation of fluid, because fluid is less echogenic than soft tissue. Many studies have demonstrated the usefulness of ultrasound for this indication. Pleural effusions as small as 3–5 mm can be detected ultrasonographically.⁵ Clinical examination is neither sensitive nor specific for the detection of pleural effusion.⁶ Pleural ultrasonography is superior to standard chest radiography in detecting the presence of pleural effusions and in distinguishing pleural effusions from atelectasis or pleural thickening.^5,7 Compared with the reference standard of chest computerized tomography (CT) scan, pleural ultrasound has 93% sensitivity and specificity for pleural effusions.⁸ When a patient has complete opacification of a hemithorax, ultrasound has 95% sensitivity for pleural effusion.⁹

The supine chest radiograph in patients in the ICU has poor performance characteristics for the detection of pleural fluid. ICU radiographs suffer from problems with penetration, rotation, and magnification. In the supine patient, pleural effusions accumulate in dependent areas. Thoracic opacities in a supine chest radiograph may be caused by a pleural effusion, a parenchymal process, such as consolidation, or by a combination of these processes. Pleural ultrasound outperforms chest radiography when compared with chest CT for identification of pleuropulmonary abnormalities.⁸ Intensive care unit radiographs often cannot distinguish between pleural and parenchymal abnormalities.^10–12 In a series of ICU patients, supine radiographs detected only 61.4% of pleural effusions when compared with those detected by ultrasound.¹³

Free-flowing pleural effusions layer posteriorly in the thorax of the supine patient. Patients with multiple lines or a compromised hemodynamic and oxygenation status are difficult to position sitting upright in bed. If the patient is supine, the bed mattress may prevent the easy visualization of small pleural effusions. One option is for the examiner to place the transducer in the posterior axillary line while angling the probe up toward the center of the body to visualize smaller effusions. In unstable patients, who have effusions that are difficult to visualize, positioning the patient in a lateral decubitus position may be helpful. The examiner should always identify three findings (Figure 18-1) indicating the presence of a pleural effusion:

1. 
   

   Anatomic boundaries: This requires definitive identification of the diaphragm and subdiaphragmatic organs (liver and spleen, depending on the side), the heart (on the left side), the chest wall, and the surface of the lung (Figure 18-1 and Video 18-1).

   
   
2. 
   

   Hypoechoic space: This requires definitive identification of a relatively echo-free space surrounded by the typical anatomic boundaries that is the pleural effusion (Figure 18-2 and Video 18-2).

   
   
3. 
   

   Dynamic changes: This requires definitive identification of dynamic changes that are characteristic of a pleural effusion (Videos 18-3 and 18-4).