Imaging of the Chest

Published on 13/02/2015 by admin

Filed under Cardiovascular

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 1401 times

Imaging of the Chest

Myriam Riboh and Patrick Knott

Wilhelm Roentgen discovered the x-ray in 1895 and won the very first Nobel Prize in physics (1901) for this discovery.1 Since that time, radiography has been used in medicine to image the chest structures. Ionizing radiation has its disadvantages, however, as Thomas Edison learned when his laboratory assistant, Clarence Daley, became the first scientist to die of radiation exposure in the United States.1 By the 1940s, ultrasound was being used as a way to image the body using nonionizing radiation.1 Computed tomography (CT) was developed in the 1970s, and magnetic resonance imaging (MRI) in the 1980s.1 All these methods have been used to image the chest, and each offers distinct advantages and disadvantages for the patient and the clinician.

Most imaging machines still use film to capture the image, but newer systems are using digital formats. In either case, the basic methods are similar: x-rays are generated when the anode of an x-ray tube is bombarded with electrons from the cathode of the tube. The collision gives off energy in the form of x-radiation, which travels out of the tube and through the patient, then hitting the imaging cassette. The cassette contains film or a digital imaging sensor. The image is then processed by a digital processor or a film developer (Figure 11-1).

The film turns black (is exposed) if the x-rays pass through the patient and reach the film surface. If the x-rays are reflected or absorbed by the patient, they do not reach the film, and the resulting image stays white (unexposed). The patterns and shades of light and dark on the film reflect the differing densities present in the part of the human body being imaged.

The darkest images on a film (also called radiolucent areas) represent pockets of air within the body. Fat is denser than air, and it produces a dark-gray image. Muscle and other soft tissues are more dense and produce a much lighter gray image. Finally, bone is the most dense natural substance in the body and produces a white image. Metallic objects are even more dense (also called radiopaque) than bone and produce a pure white image (Figure 11-2).

Figure 11-2, A, demonstrates all the different densities. This lateral view of the knee shows metal from a total knee replacement, along with bone, muscle, fat, and on either side of the patient’s knee, air.2

The x-ray image is a summation of all the densities that the x-rays have passed through. The different layers of tissue that are on top of one another are flattened into a single two-dimensional image. Sometimes the various densities lie next to one another and are easily distinguished in the image. Other times the two densities overlap one another and are blurred together in the image. For this reason, the patient is usually positioned so that two or more images can be taken at a right angle to one another. This allows structures that are overlapping in one orientation to be seen side-by-side with the other orientation.1,3

The Radiograph and Its Evaluation

The process for evaluating a radiograph can be broken down into several stages. The initial evaluation should verify all of the technical details of the scan; subsequent stages look at various aspects of the anatomy imaged.

Initial Evaluation

The initial stage is when the technical details of the images are noted:

Before proceeding with any further evaluation, the first step is to check that the information identifying the patient has been properly provided on the image and that the image is indeed of the patient in question. Identifying information usually consists of the patient’s full name, date of birth, and if possible, their sex and social security number. Failing to verify this basic information can lead to medical errors in one or multiple patients.

Next, check that the part of the body in question is indeed the part that was imaged and that it was done in at least two projections oriented 90 degrees apart (for example, one view from the front and one from the side). The left and right sides of the patient should be indicated on the films so that they can be viewed in proper orientation. A marker should also indicate the patient’s position at the time of the radiograph. For instance, a frontal chest radiograph can be taken with the patient standing, sitting, or lying down.

Finally, check the overall exposure of the film. If all the structures are too dark, the film has been overpenetrated. It should be retaken using less radiation. If all the structures are too light, the film has been underpenetrated and should be retaken using more radiation.

Evaluation for Pathology

Overview

After verifying the technical aspects of the images, the diagnostic evaluation can begin. It will help to first review the different projections that have been taken (Table 11-1). There are special views that position the patient in such a way that a particular area of interest is oriented for optimal visualization.

Table 11-1

Standard Radiograph Projections

View Description
AP (anteroposterior) Frontal view, taken with the patient facing the x-ray tube, with the back toward the film
PA (posteroanterior) Looks the same as the AP, but the patient is positioned facing the film
Right lateral Side view, with the right side against the film
Left lateral Side view, with the left side against the film
Oblique A view taken at a 45-degree angle, between the AP and the lateral views

The standard chest radiograph is taken as a posteroanterior (PA) and a left lateral view, with the patient standing. These two views position the heart closest to the film; clinicians use the radiograph to judge the size of the heart, and the farther away from the film the heart is, the more it will appear magnified. This concept is easy to understand if you shine a light on your hand and view the resulting shadow on the wall. The closer your hand is to the wall, the more its shadow’s size is true to life. The farther away the hand gets from the wall, the larger and fuzzier the shadow appears. The evaluation of the heart and lungs is most accurate when these structures are positioned close to the film during imaging.

The next step in evaluating a chest radiograph is to do a quick scan of the entire image to look for obvious or life-threatening abnormalities. General anatomy should be identified, including:

At this point it is also important to assess the position of the patient by looking at the clavicles to ensure they are aligned. If they are rotated, this could distort images of the lungs, heart, and other important structures. The image should also be scanned for any nonanatomical findings such as pacemakers, tubes, electrical leads, staples, etc. If old films exist in the patient record, it is helpful to compare the current findings with the old. Once this is done, a more thorough evaluation is begun (Table 11-2).

Table 11-2

Summary of Chest X-Ray Evaluation

Focus Findings
Patient information Name, date of birth, social security number, sex, existence of any old films.
Nonanatomical findings Pacemakers, ECG leads, buttons, surgical remnants (e.g., staples), foreign bodies.
General inspection Check rotation of chest by evaluation of the clavicles. Does anything “jump out” at you? Is everything present that should be there?
Bones Are the appropriate number present? Are they in normal anatomical position? Any fractures, lesions?
Trachea Is it midline? Is it narrowed? Are there any foreign bodies?
Lungs Assess inspiratory effort, diaphragms, lung tissue, full expansion of lungs, opacities, infiltrates.
Heart Size, borders, shape.
Mediastinum Is it widened?

Bony Skeleton

The bony skeleton (Figure 11-3) should be checked for signs of fracture, especially along the margins of the ribs. Fractures are seen as radiolucent lines within the bone or as disruptions of the smooth cortical line at the edges of the bone. Unexplained fragments of bone that are not in a normal anatomical position are another sign of fracture.

Lung Fields

Next, the lung fields should be evaluated for uniform density and vascular markings throughout. The hilar area, where the vascular and airway structures converge, is normally more dense and can be an area of importance when looking for masses or tumors. The peripheral lung fields, where the smallest vascular and airway structures are, appear the smoothest and least dense. Comparing them side by side for symmetry helps in the recognition of abnormal areas. Conditions such as pneumonia, tumor, effusion, or other disease processes lead to consolidation of lung tissue, making it more dense than tissue in the rest of the lung fields (Figure 11-4).

Full expansion of the lung fields is also important to evaluate. A pneumothorax causes part of the lung to collapse, creating a less dense empty space in the chest cavity that is devoid of any pulmonary or vascular markings (Figure 11-5). Conversely, a hemothorax or pleural effusion fills the chest cavity with fluid that is denser than normal lung tissue (Figure 11-6). Both of these abnormalities are evident on chest radiographs and can affect oxygenation, as discussed above.

Other Imaging Methods

Computed Tomography

CT scans of the chest can be a useful diagnostic tool for certain diseases (Figure 11-8). Like standard radiographs, CT uses ionizing radiation, but it allows for rapid scanning in much more detail. The CT scan generates pictures that are “slices” through the patient in the axial plane (Figure 11-9). Although bone is visualized much better than soft tissues, a CT scan can still provide excellent images of the lungs, vascular structures, and heart. New software can produce three-dimensional images that show the anatomy beautifully (Figure 11-10), and the sensitivity of these machines can allow for the detection of very small amounts of calcification in the coronary arteries, which can be an indication of early heart disease. Newer “ultrafast” CT scanners can see more subtle changes in the coronary arteries because they take images so quickly that they seem to “stop” or “freeze” the action and movement of the heart. This results in clearer pictures and the ability to discern small details more accurately.4 Another modality, called spiral CT scan, provides higher resolution and more rapid images, making it very helpful in the diagnosis of pulmonary embolism, because it can visualize vessels with great precision. In certain cases, an iodine-based dye must be injected intravenously to enhance CT images. This dye can exacerbate renal disorders, and should thus be used with caution.

As CT scans become more and more precise, they are also used more frequently. This increased use raises a concern regarding the radiation exposure of patients undergoing these studies. Although radiation exposure is also a factor with conventional radiography of the chest, the level of exposure with computed tomography has been documented to be much higher. A typical chest CT will expose a patient to the radiation equivalent of 50 to 450 chest x-rays. This increase in radiation exposure had been linked to cancer development later in life, most notably in the pediatric population. Because of this risk, other chest imaging techniques may be considered if appropriate.5,6

Magnetic Resonance Imaging

Magnetic resonance imaging of the chest can be performed to evaluate the soft tissues of the chest cavity.7 MRI has the advantage of using magnetic energy rather than ionizing radiation, so there is no harmful exposure for the patient. MRI scans also provide much higher-quality views of soft tissues than can be achieved by CT scans, but MRI is more expensive and takes nearly an hour to complete. A chest CT can be done in less than a minute, making it a better choice for evaluation of trauma (Figure 11-11). MRI is most useful when evaluating heart function/wall movement, vasculature of the heart and lungs, tumor size and staging, and respiratory movement of the muscles and diaphragm.4,8

Radionuclide Ventilation-Perfusion Scan

A radionuclide ventilation-perfusion (VQ) scan can be used to detect conditions that affect blood flow in the lung parenchyma. Radioactive isotopes are injected into the blood stream and are also inhaled into the lungs. Scanning for the location of these isotopes can then determine the extent to which the lungs have been ventilated and perfused. A mismatch (for instance, an area that was perfused with gas but not perfused with blood) would indicate that a blood clot had become caught in the pulmonary vasculature. Research has found, however, that VQ scans produce a high number of indeterminate results, and newer CT images have been found to be more accurate. A pulmonary angiogram is still the definitive test for a pulmonary embolism, but it is an invasive test that must be administered in the cardiac catheterization laboratory and involves injection of dye into the arterial vasculature; thus other imaging methods should be considered when appropriate.

Positron Emission Tomography Scan

Positron emission tomography (PET) scanning is a newer form of radionuclide imaging that helps show the function of organs rather than just their structure. For instance, the amount of radio-labeled glucose taken up by heart muscle cells can help clinicians determine whether the muscles have died after a myocardial infarction or whether they are still alive. This makes the PET scan a crucial decision-making tool: if the post-MI heart muscle cells are alive, they would benefit from bypass surgery, whereas if they are dead, bypass surgery would not be worthwhile. This modality is also very helpful in the localization of tumors in the chest cavity; these tumors are highly metabolic and have high uptake of the radionuclide, thus “lighting up” on a PET scan.9