Chapter 5 Imaging the Chest

The Chest Radiograph

Joshua Broder, MD, FACEP

Chest radiograph or x-ray is one of the most commonly performed imaging tests. It is a high-yield test, providing significant clinical information rapidly, at low cost, and with low radiation exposure, but many examinations are nonetheless unnecessary. In this chapter, we discuss the indications for chest x-ray, features of the chest x-ray technique and their effect on the resulting image, basic principles of chest x-ray interpretation, and a systematic algorithm for image interpretation. Following this, we present extensive figures with chest x-ray images depicting a spectrum of emergency findings. When available, corresponding computed tomography (CT) images are shown. These clarify chest x-ray findings, demonstrate the limitations of plain chest radiography, and illustrate the occasional need for advanced imaging for further delineation of pathology. This chapter focuses on chest x-ray in the nontrauma patient, although the principles also apply to patients with traumatic injuries, discussed in detail in Chapter 6.

Chest CT, ventilation–perfusion (VQ) scan, and thoracic ultrasound are discussed in detail in additional chapters focused on chest trauma (see Chapter 6), pulmonary embolism and aortic disease (see Chapter 7), and cardiac imaging (see Chapter 8).

Chest Imaging Modalities

Modalities for thoracic imaging include chest radiograph (x-ray), CT, VQ scan, echocardiography, magnetic resonance (MR), fluoroscopy (used for pulmonary angiography, aortography, and upper gastrointestinal imaging), and positron emission tomography (PET). MR has a limited role in the emergency department currently, largely because of issues of cost and availability, but is used in selected patients in diagnosis of conditions such as pulmonary embolism and aortic disease. PET scanning is used primarily in detection of metabolically active neoplastic disease and is therefore not routinely used in the emergency department. Characteristics of these imaging modalities are shown in Table 5-1. Issues related to CT and MR such as iodinated contrast nephropathy and allergy, radiation, and gadolinium-associated nephrogenic systemic fibrosis are discussed in Chapters 6, 7, 8, and 15.

Cost of Chest X-ray

Chest x-ray is one of the most cost-effective imaging examinations, with the cost to patients being between $50 and $220 in many medical systems.1 However, charges for chest x-ray vary considerably, even in the same geographic region. According to data provided by health care facilities, the U.S. national average cost of chest x-ray was $370 in 2010.2 Evidence-based guidelines for imaging should be followed when available and applicable, as the total cost of common, individually inexpensive tests such as chest radiography to the U.S. medical system is enormous.

Radiation Exposure From Chest X-ray

Chest x-ray provides minimal radiation exposure, around 0.01 to 0.02 mSv.3 This is the equivalent of approximately a single day’s background radiation exposure at most locations on the Earth. It equals the additional radiation exposure incurred by taking a commercial airline flight at 35,000 feet from New York to Los Angeles. The radiation exposure is focused and is quite safe in pregnant patients, with negligible scatter radiation to the abdomen and pelvis. Abdominal shielding has not been shown to be necessary to reduce fetal exposure substantially, though it remains a recommended action for the psychological well-being of the pregnant patient.4-5 Radiation exposures from chest CT are far greater, by as much as 500 fold, and are discussed in detail in Chapters 7 and 8.

Are Health Care Workers at Risk From Chest X-ray Radiation Exposure?

The exposure to health care workers from portable chest x-rays is very low if simple precautions are taken; at a distance of 15 inches from the x-ray beam, the scatter radiation exposure is minimal (around 0.0008 mSv), requiring exposure to more than 1200 radiographs to equal annual background radiation exposure.6 Multiple other studies suggest trivial exposures to health care workers using simple precautions to distance themselves from the x-ray source.6-9 If an appropriate safe distance cannot be achieved because of patient care requirements, simple shielding devices such as thyroid and torso aprons should be used.

Clinical Indications for Chest X-ray

For reasons of time, cost, and radiation exposure as described earlier, best-evidence guidelines should be applied when possible to guide the use of chest x-ray. Chest x-ray is useful in evaluation of a multitude of patient presentations, including cough and fever, chest pain, dyspnea, hemoptysis, weight loss, hypoxia, suspected subdiaphragmatic pneumoperitoneum, even Horner’s syndrome. Given the incredible range of possible patient presentations, no single clinical decision rule can govern chest x-ray use for all cases, and the emergency physician must make clinical decisions for radiography based on the individual patient presentation. However, the value of chest x-ray in common clinical scenarios has been reviewed extensively by the American College of Radiology (ACR) in its appropriateness criteria. These criteria, published on the ACR website, are expert consensus guidelines based on review of medical literature by radiologists and experts from other medical specialties. Summaries of ACR guidelines relevant to thoracic imaging are presented in Tables 5-2 and 5-3. The ACR rates imaging studies using a 1-to-9 scale, with 9 being “most appropriate” and 1 being “least appropriate.” Although the use of plain chest radiography may appear “harmless” given its speed, relatively low cost and low radiation exposure, collective costs of unnecessary radiographs and other preoperative tests are believed to total in the billions of dollars in the United States annually.10 Other potential harms from unnecessary radiography include delays in surgery prompted by the detection of incidental findings such as pulmonary nodules and costs and morbidity of additional diagnostic procedures triggered by chest x-ray findings. Consequently, chest x-ray should be ordered with attention to the evidence supporting its use for a given clinical presentation. Unfortunately, as we will review, high quality evidence to guide clinical decisions is lacking for many potential indications for chest x-ray.

TABLE 5-2 American College of Radiology Appropriateness Criteria for Chest X-ray or Computed Tomography in Common Clinical Scenarios

Clinical Scenario ACR Rating for X-ray Chest ACR Rating for CT Chest
Acute chest pain    
9 CTA chest (noncoronary): 6—Useful in ruling out other causes of chest pain such as aortic dissection, pulmonary embolism, pneumothorax, pneumonia.
9—Should be performed if readily available at the bedside and does not cause delay in obtaining a CT or MRI. Alternative causes of chest pain may be discovered. Not the definitive test for aortic dissection. CTA chest and abdomen: 9—Recommended as the definitive test in most patients with suspicion of aortic dissection.
9—to exclude other causes of chest pain CTA chest (noncoronary): 9—Current standard of care for detection of PE.
Chronic chest pain    
3—Usual initial imaging study in cardiac patients. Although used frequently, chest radiographs can neither establish nor exclude chronic ischemic heart disease. Insensitive for detecting coronary arterial calcification. Limited value in patients with high risk of CAD. CTA coronary arteries: 7—Very good accuracy and negative predictive value in low to intermediate risk groups. However, may have false negatives in high-risk group, and negative studies may still require further diagnostic testing. Coronary calcification often found in older high-risk patients (especially males) can limit coronary luminal assessment.
9 CTA chest (noncoronary) with contrast: 8— Important examination for pulmonary embolism and thoracic aortic aneurysm/dissection. To rule out PE and evaluate lung pathology. Appropriate for chronic anginal chest pain.CTA coronary arteries with contrast: 8—Can be used to assess for coronary atherosclerosis,anomalous coronary artery, and pericardial disease. High negative predictive value will exclude coronary artery disease and allow triage management to focus on more likely diagnoses. To eliminate unnecessary catheterizations.
CHF    
9 2—readily diagnosed on CT obtained for other indications
4
Dyspnea—suspected cardiac origin 8 CTA coronary arteries: 6CTA coronary arteries with advanced low dose techniques: 6CTA chest (noncoronary): 6
Suspected bacterial endocarditis 9 CT heart function and morphology with contrast: 6—Multidetector with maximal temporal and spatial resolution. Probably indicated to rule out paravalvular abscess and/or psedoaneurysm. Emerging technology.
Known or Suspected congenital heart disease in the adult 7 CT heart function and morphology with contrast: 7. May be alternative to MRI and TTE/TEE. High spatial resolution to evaluate small and tortuous vessels. For evaluation of airway. CTA coronary arteries with contrast: 6. When there is a high index of suspicion for fistula or anomalous coronaries.CTA coronary arteries with contrast with advanced low dose techniques: 6. Important in young to middle-age adults because of reduced radiation dose.
Acute respiratory illness Varies—see Table 5-3 Varies
Acute respiratory illness in HIV-positive patients 9 CT chest without contrast: 8—When negative, equivocal, or nonspecific chest radiograph.
Chronic dyspnea—suspected pulmonary origin 8-9—depends on age and examination Any age, nonrevealing or nondiagnostic clinical, standard radiography, and laboratory studies. In the setting of chronic dyspnea, the most appropriate imaging study is a thin section high resolution chest CT: 9. If a patient has dyspnea not clearly of pulmonary origin, other entities such as chronic or acute pulmonary embolism may need to be excluded. In that setting, a thin section chest CT with intravenous contrast is appropriate. See the ACR Appropriateness Criteria® topic on “Acute Chest Pain–Suspected Pulmonary Embolism”.
Hemoptysis,Two risk factors (>40 years old and >40 pack-year history) 9 without contrast: 6 (useful for patients with renal failure or contrast allergy.); with contrast: 8 (optimal study shows enhancement of the systemic arteries.)
Rib fractures, adult < 65 years of age 8—PA view. Rib views are given a rating of 2. 2—with or without contrast
Routine admission and preoperative chest radiography   NA
2 (preoperative or routine)
9 (preoperative or routine)
6 (preoperative), 4 (routine admission)
8 (preoperative or routine)
Routine chest radiographs in uncomplicated hypertension (asymptomatic)   NA
1 NA
5 NA
Screening for pulmonary metastases 8-9—depends on malignancy type 7-9—depends on malignancy type; noncontrast CT
Solitary pulmonary nodule NA—chest x-ray presumably already performed, demonstrating lesion CT chest with contrast: 6CT chest without contrast: 8
Staging of bronchogenic carcinoma CT chest with or without contrast (including upper abdomen): 9—CT with contrast is preferred if there are no strong contraindications.
Blunt chest trauma—suspected aortic injury 9

This table integrates information relevant to the practice of emergency medicine from multiple ACR Appropriateness Criteria topics, and does not represent all information included in these topics. Refer to the ACR website at www.acr.org/ac for the most current and complete version of the ACR Appropriateness Criteria®.”

Reprinted with permission of the American College of Radiology, Reston, VA. No other representation of this material is authorized without expressed, written permission from the American College of Radiology. Rating scale: 1, 2, 3: Usually not appropriate; 4, 5, 6: may be appropriate; 7, 8, 9: Usually appropriate.

From American College of Radiology. ACR appropriateness criteria. (Accessed at http://www.acr.org/secondarymainmenucategories/quality_safety/app_criteria.aspx) (Access date: 3-21-2011)

TABLE 5-3 American College of Radiology Appropriateness Criteria: Acute Respiratory Illness

Clinical Variant ACR Rating for X-ray Chest
>Age 40 8
Dementia, any age 8
<Age 40, negative physical exam, and no other signs, symptoms, or risk factors 4
<Age 40, positive physical exam or other risk factors 9
Complicated pneumonia 9
Suspected severe acute respiratory syndrome 9
Suspected anthrax 9
Febrile, neutropenic 9
Acute asthma  
4
9
Acute exacerbation of COPD  
4
9

This is a truncated version of the full ACR Appropriateness Criteria, which also includes ratings for chest CT in each of the Clinical Variants.Reprinted with permission of the American College of Radiology, Reston, VA. No other representation of this material is authorized without expressed, written permission from the American College of Radiology. Refer to the ACR website at www.acr.org/ac for the most current and complete version of the ACR Appropriateness Criteria.

Acute respiratory illness is defined by the presence of one or more factors including cough, sputum production, chest pain, and dyspnea, with or without fever.

American College of Radiology: ACR Appropriateness Criteria: Acute Respiratory Illness. Available at: http://www.acr.org/SecondaryMainMenuCategories/quality_safety/app_criteria/pdf/ExpertPanelonThoracicImaging/AcuteRespiratoryIllnessDoc1.aspx. Accessed 3-21-2011. Rating Scale: 1,2,3 Usually not appropriate; 4,5,6 May be appropriate; 7,8,9 Usually appropriate

Is a Chest X-ray Routinely Indicated for Chest Pain?

A multitude of causes can result in chest pain, including a number of abnormalities visible on chest x-ray. Pneumonia, pneumothorax, pneumomediastinum, neoplasms, rib fractures, and aortic dissection are just a few of the disease processes that can result in chest pain and that the chest x-ray may diagnose or suggest. Pulmonary embolism may have associated chest x-ray abnormalities such as pleural effusion or pulmonary infarct, although sensitivity and specificity of chest x-ray are too poor for clinical diagnosis of this entity. The ACR recommends chest radiography as the initial imaging study for patients with chest pain and a low probability of cardiac ischemia, given the range of possible diagnoses and the low cost and radiation exposure of chest x-ray.11 The American College of Emergency Physicians (ACEP) recommended chest x-ray for many scenarios described in its 1995 “Clinical Policy for the Initial Approach to Adults Presenting With a Chief Complaint of Chest Pain, With No History of Trauma,” which has since been retired (all ACEP clinical policies are subject to retirement if a new review of the literature is not conducted within a prescribed time frame).12 Systematic prospective studies on the value of chest x-ray in undifferentiated chest pain are limited. In a retrospective study of 5000 portable or posterior–anterior (PA) and lateral chest radiographs performed on emergency department patients, Buenger et al.13 discovered that “serious disease” was found in 35% of those with chest symptoms, based on the x-ray requisition. Requisition information may not reliably characterize a patient’s symptoms, so the validity of this study remains in doubt.

Is a Chest X-ray Routinely Indicated for Stroke?

Sagar et al.14 performed a retrospective review of 435 patients admitted with a diagnosis of acute stroke. They found that 86% had an admission chest x-ray performed and 77.5% of radiographs were technically inadequate, mostly because of poor patient positioning and poor inspiratory effort. Radiographic abnormalities were seen in 61 patients (16.4% of those undergoing radiography), and clinical management appeared to have been altered in 14 patients (3.8% of those undergoing radiography). The authors conclude that chest radiography is not routinely indicated for patients with acute stroke in the absence of suggestive clinical findings or complaints. As with many other clinical presentations, prospective studies for chest x-ray in stroke are lacking, and little evidence exists regarding the role of chest x-ray in presentations such as altered mental status (rather than obvious stroke syndromes such as hemiparesis).

Is a Chest X-ray Routinely Indicated for Gastrointestinal Bleeding?

Tobin et al.15 reviewed 202 adult patients admitted to intensive care units for gastrointestinal hemorrhage. They found that 161 (80%) underwent admission chest x-ray. Review of radiographs by a radiologist blinded to the study purpose found 14.3% with minor radiographic abnormalities, none of which resulted in a therapeutic or diagnostic intervention. Major radiographic abnormalities were found in 13% of patients, with an intervention in only 5.6% of patients. Predictors of major abnormalities included history of lung disease and abnormal lung physical examination at presentation (sensitivity = 79%, negative predictive value = 96%). These variables also predicted major radiographic abnormalities associated with interventions (sensitivity 89%, negative predictive value = 99%). The authors found that selective radiography based on these criteria would have reduced the charges from chest radiography by more than $500 per major abnormality detected and by more than $1000 per major abnormality associated with an intervention. These findings have not been externally validated in a separate population but do suggest that routine chest x-ray is unnecessary for this clinical presentation.

Is a Chest X-ray Routinely Indicated for Acute Respiratory Illness With Features Such as Productive Cough?

Acute respiratory illness is defined by the presence of one or more factors, including cough, sputum production, chest pain, and dyspnea, with or without fever. A multitude of studies have examined the role of chest x-ray, examining varying populations including healthy hosts, immunocompromised patients, the elderly, and patients with asthma and chronic obstructive pulmonary disease (COPD). The ACR divides its recommendations based on these and other patient factors as shown in Table 5-3.17 Benacerraf et al.16 studied 1102 outpatients with acute respiratory illness and found that in patients below 40 years 96% of chest radiographs had no acute abnormalities. Limiting radiography in patients below age 40 to those with abnormal physical examination or hemoptysis would have reduced radiography by 58% and missed only 2.3% of acute abnormalities. Fever, cough, and the presence of even purulent sputum were not predictive of chest x-ray abnormalities in patients under the age of 40.16-17 Patients above the age of 40 and immunocompromised patients, such as those with HIV or neutropenia, represent higher risk groups for which radiography is recommended by the ACR for signs and symptoms of acute respiratory illness.17 The emergency physician should incorporate other factors into the decision to obtain radiography and to treat the patient, including access to care, contacts with high-risk (immunocompromised) patients, endemic or epidemic disease, and risk factors for tuberculosis or undiagnosed HIV.

Is a Chest X-ray Routinely Indicated for Suspected Pneumonia?

When the emergency physician specifically suspects pneumonia, is radiography warranted? As discussed earlier, patients below the age of 40 with normal physical examination and no hemoptysis have a very low rate of radiographic abnormalities and should not routinely undergo chest x-ray. Does this mean that patients not meeting these “low risk” criteria must undergo chest x-ray? The ACR points out in its appropriateness criteria for acute respiratory illness some important disparities between clinical practice and professional society recommendations in the diagnosis of pneumonia. The Infectious Diseases Society of America and the American Thoracic Society recommend that chest radiography be obtained when pneumonia is suspected in adults to assist in distinguishing community-acquired pneumonia from other common causes of cough and fever, such as acute bronchitis. However, some clinicians may choose to treat patients empirically for pneumonia, without obtaining radiography, when a high clinical suspicion exists. This is an appropriate evidence-based strategy, as a negative chest x-ray does not rule out pneumonia. When a high pretest probability exists, a negative chest x-ray is inadequate to exclude pneumonia and should not be used as the sole basis to determine management.17

Is a Chest X-ray Routinely Indicated for Asthma?

In patients with asthma without features suggesting complications such as pneumonia or pneumothorax, x-ray likely has little role.17 Findley and Sahn18 studied 90 adult patients with asthma undergoing chest radiograph and found only 1 patient with an acute infiltrate; normal (55%), hyperinflation (37%), and unchanged minimal interstitial infiltrates (7%) were common patterns. Chest x-ray interpretation did not correlate with admission, suggesting that radiographs in asthma play little role in management. Gershel et al.19 studied 371 consecutive children over the age of 1 year with first-time wheezing and found that 94.3% had x-ray findings compatible with uncomplicated asthma (negative). Of the 371 patients studied, 21 (5.7%) had positive chest x-ray findings, including atelectasis and pneumonia in 7, segmental atelectasis in 6, pneumonia in 5, multiple areas of subsegmental atelectasis in 2, and pneumomediastinum in 1. Although the authors call this 5.7% “positive,” it is not clear that findings of atelectasis or even pneumomediastinum alter management in a clinically beneficial way, as no specific therapy is required for these abnormalities. When only pneumonia is considered a positive finding, only 3.9% of first-time wheezing children had positive chest x-rays. The authors found that 95% of abnormal x-rays could be predicted based on the presence of tachypnea greater than 60 breaths per minute, pulse greater than 160 beats per minute, and localized rales or diminished breath sounds, before or after treatment—suggesting that even in first-time pediatric wheezing, routine x-ray may not be necessary. Aronson et al.20 reported on 125 consecutive adult asthma admissions and found no chest x-ray abnormalities in 81 patients characterized as uncomplicated, defined as having a prior history of asthma, presenting with dyspnea and wheezing, and having no history of fever or chills, immunosuppression, cancer, intravenous (IV) drug abuse, prior thoracic surgery, cardiac disease, or pulmonary disease such as COPD or granulomatous disease.

Is a Chest X-ray Routinely Indicated for Chronic Obstructive Pulmonary Disease?

In COPD, the ACR recommendations are somewhat at odds with work published in the emergency medicine literature. According to the ACR, in COPD patients without fever, leukocytosis, chest pain, or history of congestive heart failure (CHF) or coronary artery disease (CAD), chest x-ray has a low diagnostic utility, with an appropriateness rating of four out of nine.17 These recommendations are concordant with the work of Sherman et al., who retrospectively reviewed 242 patients with admission chest x-rays for COPD. Although 14% had radiographic abnormalities, the authors concluded that only 4.5% had appropriate and clinically important management changes resulting from chest x-ray findings. They found that leukocytosis (white blood cell count >15 × 109 per liter and neutrophil count >8 × 109 per liter), history of CHF or CAD, chest pain, and extremity edema were a “high risk” criteria predictive of important acute radiographic abnormalities.

Emerman and Cydulka21 attempted to validate these high-risk criteria in a retrospective series of 685 emergency department visits for acute COPD in which a chest x-ray was performed. In this series, 16% of patients had significant chest x-ray abnormalities: 88 with new infiltrates, 2 with new lung masses, 1 with pneumothorax, and 20 with evidence of pulmonary edema. Abnormalities were associated with history of CHF, fever, rales, pedal edema, and jugular venous distension. In contrast, white blood cell count, temperature in the emergency department, history of CAD, and complaints of chest pain or sputum production were not associated with chest x-ray abnormalities. The sensitivity of the high-risk criteria was only 76%, with a specificity of only 41%. The authors recommended routine x-ray in patients with apparent acute COPD.

Is a Chest X-ray Routinely Indicated for Asymptomatic Hypertension?

According to the ACR reviews, patients with asymptomatic hypertension should not routinely undergo chest x-ray, although debate remains about the value of radiography with rising levels of hypertension. In theory, chest x-ray might be of value in patients with asymptomatic hypertension by revealing cardiomegaly or left ventricular hypertrophy, aortic enlargement or coarctation, or pulmonary vascular congestion. Although a number of authors have suggested these benefits, others have argued that the management of hypertension is not positively influenced by chest x-ray findings; regardless of the presence or absence of chest x-ray abnormalities, blood pressure control should be attempted. Moreover, studies indicate that chest x-ray is insensitive and nonspecific for detection of left ventricular hypertrophy, which is better assessed with echocardiography.22

Is a Chest X-ray Routinely Indicated for Admitted Patients?

The ACR found that routine admission or preoperative chest x-ray is not indicated in the patient with no cardiopulmonary findings by history or physical examination, with an appropriateness of two out of nine.10 In patients with acute cardiopulmonary findings by history or physical, chest x-ray is highly appropriate, with a rating of nine. Because of a lack of evidence against the use of routine chest x-ray, in elderly patients (70 years or older) with chronic cardiopulmonary disease, the ACR rates chest x-ray as appropriate (eight of nine), even in the absence of history or physical findings, when a chest x-ray has not been performed within the past 6 months. The value of chest x-ray is less clear in this asymptomatic group when a recent chest x-ray (within 6 months) is available for review (ACR rating four to six of nine).

What is the basis for the ACR recommendations in these cases? The ACR cites multiple studies addressing the utility of routine chest x-ray in patients admitted for gastrointestinal hemorrhage, acute stroke, and other conditions affecting the elderly. Gupta, Gibbins, and Sen23 prospectively studied 1000 consecutive admissions to an acute geriatric ward and found that 35% to 50% of patients had no apparent indication for chest x-ray. In this group without chest x-ray indications, 5.5% had radiographic abnormalities but less than 1% had clinically important findings. In an exhaustive systematic review, Munro, Booth, and Nicholl24 found that no randomized controlled trials of the effectiveness of routine preadmission or preoperative chest x-rays had been published, and most studies do not (or cannot, because of study design) assess the effect on clinical management of detected radiographic abnormalities. Estimates of the frequency of changes in clinical management suggest patient benefit in fewer than 5% of cases.10

We have discussed some common evidence-based indications for chest x-ray. Now we consider basic principles of chest x-ray technique relevant to emergency medicine and key principles of image interpretation. We finish with findings of specific pathology and a systematic approach to the interpretation of chest x-ray.

Interpretation of Chest X-ray: Basic Principles

Interpretation of chest x-ray requires several elements:

We start with some basic principles. If you are already familiar with basic radiographic concepts and the normal appearance of the chest x-ray, move on to the section on pathology. But a few minutes familiarizing yourself with the ideas in this section can pay big dividends some day when you encounter an unfamiliar chest x-ray that does not fit any of the patterns of common pathology discussed in this chapter. Understanding basic principles may allow you to deduce the nature of the abnormality and avoid common pitfalls. A number of outstanding texts are devoted entirely to chest radiography.25-26 We have drawn on numerous sources for the following discussion.

Chest X-ray Techniques and Effects on Resulting Images

Familiarity with the chest x-ray technique is important for the emergency physician because it can affect the appearance of important diagnostic findings and the sensitivity and specificity of a chest x-ray for these diagnoses. In some cases, a given chest x-ray technique may falsely simulate pathology; in other cases, the technique many hide important abnormalities. Here, we describe the most common chest x-ray techniques, along with pitfalls of each. The most common variations of the frontal chest x-ray used in the emergency department are the anterior–posterior (AP) view and the PA view. The AP view is commonly performed as a portable study in the patient’s room, and it may be performed with the patient in a supine or upright position. The PA frontal view is performed in the radiology suite, and often a lateral projection PA chest x-ray is obtained at the same time.

Chest x-rays can be characterized by the following:

Each of these variables has important effects on the resulting image. In some cases, the technique is manipulated intentionally to achieve a desired diagnostic effect. In other cases, the patient’s clinical condition limits the x-ray technique, and we are forced to accept a suboptimal diagnostic image. Understanding the effects of the x-ray technique on the resulting image is invaluable in helping the emergency physician to avoid misinterpretation of the image and misdiagnosis of the patient.

Direction of the X-ray Beam With Respect to the Detector: Posterior–Anterior Versus Anterior–Posterior Technique

When a frontal projection chest x-ray is obtained, the direction of the x-ray beam as it passes through the patient to the detector can be from patient posterior to anterior or from patient anterior to posterior (Figure 5-1). To restate this another way:

In the case of the PA x-ray, the patient is oriented with the x-ray film or detector in contact with the anterior surface of the thorax. The x-ray beam passes through the patient’s posterior thorax to the detector. The patient faces toward the detector, away from the x-ray source (Figure 5-1).

In the case of the AP x-ray, the patient is oriented with the x-ray film or detector, in contact with the posterior surface of the thorax. The x-ray beam passes through the patient’s anterior thorax to the detector. The patient faces toward the x-ray source, away from the detector.

Why does the direction of the x-ray beam through the patient to the x-ray detector matter? We return to this question in a moment, but first we describe a characteristic difference in the PA and AP x-ray techniques, which contributes to differences in the resulting image.

The PA chest x-ray is typically acquired in a radiology suite, with the x-ray source positioned 6 feet from the x-ray detector. In comparison, the AP chest x-ray is typically acquired as a portable examination, with only 3 feet separating the x-ray source and the detector.

Distance of the X-ray Beam With Respect to the Detector: Posterior–Anterior Versus Anterior–Posterior Technique

Why does the distance from the x-ray source to the detector matter? The answer becomes obvious if we consider the analogy of the children’s game of creating shadows on a wall with a light source (Figures 5-2 and 5-3). You can try our example at home (use a light, not an x-ray) if it is unclear. Place a lamp 6 feet from a wall. Position your hand close to the lamp, and the shadow on the wall appears larger than the actual size of your hand, though rather indistinct at its edges. Most people have used this trick for their amusement, making a hand appear as large as a head, for example. Without moving the lamp, move your hand closer to the wall; the resulting shadow becomes smaller (closer to its actual size), denser (darker), and sharper at its edges. We have now defined one of the two variables: with a light (or x-ray) source a fixed distance from a detector, positioning the object to be imaged close to the detector results in a sharper, truer image without false magnification. Therefore, when performing medical imaging, we should place the body part to be imaged as close to the detector as possible to achieve a sharp, unmagnified image. Now consider the same scenario, with a twist. Imagine that you cannot put your hand close to the wall because of furniture obstructing your path. How can you sharpen the image and reduce magnification? The answer is simple: move the lamp farther from your hand and the wall. Try this experiment, and you will find that the shadow of your hand becomes sharper and less magnified, truer to its actual size. Why would this scenario occur with medical imaging? Imagine placing your thorax against the wall; your ribs prevent you from moving your internal organs closer to the wall, although by facing toward or away from the wall you can position your heart (anterior in your chest) closer to the wall. However, you can easily control the distance of the light source from the wall to reduce magnification and improve the image sharpness. Therefore, when performing medical imaging, we should position the x-ray source as far as possible (within reason) from the body part to be imaged, to improve sharpness and reduce magnification.

Magnification: Asset or Artifact?

Although magnification may sound advantageous, it can introduce clinically deceptive artifacts if some body parts are magnified to a greater degree than others. The most clinically accurate x-ray has no magnification. Why would we not want magnification? Wouldn’t magnification potentially assist in identifying pathology? The problem again can be resolved by considering the shadow experiment. In medicine, we need an accurate estimate of the actual size of body parts and an accurate measure of their size relative to one another. Consider the example of the hand shadow and the shadow of the human head. Although it may be amusing to have your hand mimic a giant dog biting a head, this scenario of varying magnification is highly undesirable in medical imaging. We need to know whether a mass has increased in size or if the heart and mediastinum are enlarged relative to the surrounding thoracic cavity. Our imaging technique must avoid false magnification, particularly when differential magnification of objects in the same image might occur.

Hopefully, we have convincingly demonstrated how the distances from the body part to the detector (assuming a fixed light source) and from the light source to the body part (assuming a fixed distance from body part to detector) affect the resulting images. Let’s briefly consider why this is the case. Figures 5-4 through 5-6 illustrate the effect of distance on object magnification. The angle created between the x-ray source and an object’s edges is determined by the distance from the x-ray source to the object. The shorter the distance from the x-ray source to the object, the greater the angle. Greater angles lead to greater magnification; therefore, a shorter distance from the x-ray source to the imaged object leads to increased magnification. Although increased magnification may appear to be a benefit, in most clinical scenarios this can lead to a false appearance of cardiomegaly or a widened mediastinum. This is particularly a problem if very short distances are employed, as thoracic structures nearer to the x-ray source will be magnified to a greater degree than structures farther from the source. Positioning the x-ray source at a greater distance from the object to be imaged reduces this relative magnification, leading to a truer representation of the object. The AP portable examination places the x-ray source 3 feet from the patient, rather than the 6 feet usually used for a PA examination. Consequently, the AP portable chest x-ray examination typically has a greater degree of false magnification of the heart and mediastinum compared with the PA technique.

We asked earlier why the direction of the beam (AP vs. PA) matters to the resulting image. As we described in our shadow experiment earlier, sometimes we cannot fully control the location of objects relative to the x-ray detector. We cannot change the anterior location of the heart in the chest; however, by positioning the detector on the anterior aspect of the chest, we bring the heart and the detector closer together, reducing magnification of the heart. Consequently, the PA x-ray technique results in less magnification of anterior structures compared with the AP technique. Remember this when viewing AP portable chest x-rays, which are more prone to false appearance of cardiomegaly or mediastinal widening (Figure 5-7).

image

Figure 5-7 Differences between anterior–posterior (AP) and posterior–anterior (PA) chest x-rays and upright or supine positioning.

As discussed in Figures 5-2 through 5-6, a PA chest x-ray results in a less magnified appearance of the heart, whereas an AP x-ray results in false magnification that can simulate or exaggerate cardiomegaly. Upright positioning typically results in better lung expansion. Supine positioning typically results in poorer lung expansion and clumping of pulmonary vessels in poorly expanded lungs, with an appearance that can simulate or exaggerate the appearance of pulmonary edema. Supine positioning also exaggerates the mediastinal width. A, An upright PA chest x-ray. B, A supine AP chest x-ray obtained a short time later in the same patient. The cardiac silhouette is enlarged in both but appears larger in B.

Patient Positioning for Chest X-ray: Supine Versus Upright Technique

Now let’s also consider the differences among a true upright, a supine, and an intermediate lordotic x-ray (Table 5-4). Although occasionally other positions of the patient are desirable in specific clinical scenarios, for most applications, a perfectly upright patient position is desirable during chest x-ray acquisition. When the patient is positioned perfectly upright and perpendicular to the direction of the x-ray source, thoracic structures are positioned equal distances from the x-ray source, ensuring equal magnification on the resulting x-ray image. In comparison, if the patient is positioned in a lordotic position with the upper and lower thorax at different distances from the x-ray source, the magnification of superior thoracic structures will differ from that of the lower thoracic structures.

TABLE 5-4 Artifacts of Supine Chest X-ray Technique That Can Lead to Misdiagnosis

Supine Chest X-ray Finding Possible Misdiagnosis
Heart appears large Simulates cardiomegaly
Mediastinum appears wide Simulates aortic or mediastinal abnormality
Increased vascular markings in upper lung zones Simulates pulmonary edema
Poor inspiration Simulates pulmonary edema and may hide small nodules
Layering of fluid in plane of x-ray detector May prevent recognition of pleural fluid
Distribution of air to anterior chest and abdomen May prevent recognition of pneumoperitoneum or pneumothorax

When the patient is positioned in a fully upright position, fluid within structures of the chest will generally reside due to gravity in a dependent position, forming fluid levels. In comparison, if the patient is positioned supine, the horizontal plane in which fluid will spread is parallel to the x-ray film or detector beneath the patient, and no fluid levels will be seen. (see Chapter 6, Figure 6-2). The resulting appearance may be a diffuse increase in the density of the entire affected thorax, sometimes called a “veiling opacity.” This may be mistaken for an increased parenchymal density, rather than being recognized as a broadly layered pleural effusion.

Pleural fluid and fluid within collections such as lung abscesses typically are not visible on an image obtained with a supine patient but are visible on an image obtained with an upright position. In a patient positioned upright, the upper surface of fluid within the potential pleural space usually forms a curved line, higher along the lateral chest wall than at its intersection with the mediastinum. This appearance is termed the meniscus sign (Figures 5-8 and 5-9). In contrast, fluid within an air-filled cavity usually forms a straight horizontal line without a lateral meniscus. Examples include fluid within an air-filled abscess cavity or fluid within a hemopneumothorax (Figure 5-10; see also Figure 5-8).

In an upright patient, air within the peritoneal cavity collects beneath the diaphragm, making it visible on x-ray because of the contrast between the density of air and diaphragmatic soft tissue. (Figures 5-11 through 5-14). In a supine patient, air within the peritoneal cavity may collect in the midline anterior abdomen rather than in a subdiaphragmatic position and in addition will spread in the same horizontal plane as the x-ray detector beneath the patient, preventing the air from being visible. An x-ray obtained with an upright patient position is therefore more sensitive for detection of pneumoperitoneum.

image

Figure 5-11 Free air (pneumoperitoneum).

Free air (pneumoperitoneum) can be recognized on upright chest x-ray. This finding relies on the silhouette sign described later in this chapter (see Figure 5-26): air density is readily seen when in direct contact with water (soft-tissue) density. A, The normal appearance of the upright chest x-ray in the absence of pneumoperitoneum. B, The appearance of the upright chest x-ray in the presence of pneumoperitoneum. On an upright chest x-ray, normally the inferior surface of the right diaphragm is not seen, as the liver (water or soft-tissue density) is in direct contact with the inferior border of the diaphragm (also water density). When air is present within the peritoneal cavity, it may collect inferior to the right diaphragm, superior to the liver. This air may be recognized as a black line or collection, making the inferior border of the diaphragm visible. On the patient’s left side, the normal gastric air bubble can simulate intraperitoneal air, as it is normal for the stomach to contain air. Fortunately, usually the gastric wall and adjacent diaphragm are thicker than the diaphragm alone would be, allowing the normal gastric bubble to be distinguished from subdiaphragmic pneumoperitoneum. In some cases, pneumoperitoneum may extend beneath the central diaphragm, in which case the inferior border of the heart may be partially visible. Normally, the heart, diaphragm, and liver are in contact, with no visible line separating them. Compare with Figures 5-12 through 5-14.

image

Figure 5-12 Free air.

Pneumoperitoneum is a critical finding requiring recognition on chest x-ray. Occasionally, it may be an unanticipated finding on chest x-ray in a patient who cannot provide an adequate history. Remember that normally the inferior surface of the diaphragm cannot be seen, as it is contiguous with a solid organ sharing the same water density on chest x-ray: the liver on the right, the spleen on the left. On the left, the interior surface of the stomach may be shown in relief by air within it. It may be difficult to distinguish this inner surface of the stomach from the inferior surface of the diaphragm, although the diaphragm alone should be thinner than the combined thickness of diaphragm and stomach. Within the abdomen, the external surface of the bowel wall should not be seen, again because of its contiguity with other soft-tissue structures. Air within the bowel is readily seen and makes the internal surface of the bowel wall quite apparent. When pneumoperitoneum exists, the external surface of the bowel can be seen. A normal finding that may simulate this is the presence of two adjacent loops of bowel with their walls abutting. In this case, the internal surface of both walls may be seen, and it may appear that air is present on both sides of the wall of a single loop. In A (upright PA x-ray) and B (upright lateral x-ray), copious free air is present. A, Both diaphragms are outlined, and several bowel loops can be seen with air on both sides of their walls. B, The lateral x-ray also demonstrates this finding. Compare with the CT from the same patient in Figure 5-13.

image

Figure 5-13 Free air, abdominal CT.

Same patient as in Figure 5-12. A and B are shown on lung windows to highlight the contrast between air and other soft tissues. In these axial CT images, free air is seen outlining loops of bowel. Note how both sides of the bowel wall are clearly visible when contrasted with air. B, Notice that where a loop of bowel abuts the anterior abdominal wall, only the internal surface of the bowel wall is visible. This is because the bowel wall and the abdominal wall share the same soft-tissue density. On CT, air may accumulate along the anterior abdominal wall, as this is the highest point in the abdomen when the patient lies supine for CT. In contrast, in an upright patient undergoing chest x-ray, air accumulates under the diaphragm. Air is an outstanding contrast agent because of its much lower density than abdominal tissues, so no administered contrast agents are required to detect it.

Another advantage of the upright chest x-ray is that the patient typically is able to accomplish a greater degree of inspiration. With greater inspiration, blood vessels become more widely spaced, allowing other abnormalities to be recognized. Radiologists have sometimes compared this with seeing a bird in a tree, which is more easily accomplished in a tree with widely spaced branches. Small lung masses can be more easily seen as a result. A well-expanded chest on an upright view also gives a truer estimate of the heart size. Rarely, a chest x-ray obtained at end-expiration provides diagnostic benefits, which we will review later.

In comparison, a supine chest x-ray is usually characterized by higher diaphragms, with poorer lung expansion and clumping of pulmonary vessels. In addition, blood flow to the upper lung is relatively increased by gravity, and the heart appears larger than on an upright x-ray. This is due to the portable AP technique but also because of the relatively unexpanded chest, which makes the heart appear comparatively larger. In combination, these findings may mimic CHF (see Figure 5-7).

Supine and semilordotic views are often obtained in emergency department patients not by design but because patient condition prohibits the desired technique. Trauma patients, unstable medical patients, and intoxicated or neurologically disabled patients are just a few examples of patients who often undergo imaging in supine or semilordotic positions because of their inability to be positioned upright. The emergency physician must be aware of the limitations of these examinations and should recognize the patient position when interpreting the resulting images.

Lordotic and decubitus positions are sometimes obtained intentionally for specific diagnostic purposes, described later under “Additional Chest X-ray Views.”

Normal Appearance of a Chest X-ray

Recognizing pathology requires a strong familiarity with the normal appearance of the chest x-ray. Classic abnormalities are often recognized by their distinct differences from the normal chest x-ray appearance. More subtle abnormalities may be missed by an inexperienced observer, but an experienced reader may immediately recognize that the chest x-ray differs from the norm even before characterizing and articulating the abnormality. Radiologists understand that recognizing familiar patterns of normal and abnormal requires constant exposure, much like recognizing a familiar relative. This pattern recognition approach has been dubbed the “Aunt Minnie” effect for this similarity. We briefly describe the appearance of a normal chest x-ray here. In the sections that follow, we present numerous examples of important pathology. Even with several examples of each type, you will only begin the exposure required to “recognize Aunt Minnie.” Make a habit of looking at your patients’ chest x-rays, even if the radiologist has already rendered an interpretation.

Frontal (Posterior–Anterior or Anterior–Posterior) Upright Chest X-ray View

The normal frontal upright chest x-ray (Figure 5-15) has the following features:

Lateral Upright Chest X-ray View: Retrosternal Space, Retrocardiac Space, and the Spine Sign

The lateral chest x-ray provides important diagnostic information. Unfortunately, this view is usually not obtained when a portable x-ray examination is performed—another good reason to send the patient to the radiology suite for imaging if the clinical condition permits this. The lateral view (Figure 5-16) reveals the retrosternal space, which overlies the heart and mediastinum on a frontal projection. This space is usually quite lucent (black) because of the presence of a low-density epicardial fat pad and sometimes lung segments—but when occupied by a soft-tissue mass, this space may appear radiodense (white) (Figure 5-17). The lateral chest x-ray also reveals the retrocardiac space. This space normally should be quite lucent (black) (see Figure 5-16). Lower lobe pneumonias may be evident on the lateral view as an abnormally dense retrocardiac region (Figure 5-18). On the lateral view, the diaphragms usually form smooth curves descending from anterior to posterior. The space above the diaphragms is usually lucent (black), as it contains low-density lung tissue. Pleural effusions may be evident on lateral view as dense (white) layering opacities replacing the normal curve of the diaphragm in this space (see Figure 5-9). Sometimes pleural effusions form a meniscus against the posterior wall of the thorax, actually reversing the normal curve of the diaphragm. In addition, air beneath the diaphragm (pneumoperitoneum) may be visible on the lateral view (see Figure 5-14).

The thoracic spine also is visible on a lateral chest x-ray. The normal appearance of the spine is a gradually more lucent (blacker) appearance moving from cephalad to caudad (see Figure 5-16). This is not a result of decreasing spinal density but rather is a normal artifact of the examination technique. When this progressively more lucent appearance is lost, it implies the presence of an abnormal density in the retrocardiac space. This is called the spine sign and is a pathologic abnormality that can be a clue to disease. Remember that the increasing density has a differential diagnosis, including infectious infiltrate, pulmonary edema, pleural effusion, mass, and atelectasis. Other radiographic findings and the patient’s clinical presentation must be used to sort through this differential diagnosis, and additional imaging may be necessary. Nonetheless, this finding can confirm a pneumonia not seen on the frontal projection x-ray. The lateral x-ray is often neglected but is a key additional view that should be obtained whenever possible and carefully reviewed.

Additional Chest X-ray Views

Early radiologists became extremely skilled at deducing clinically relevant information from additional chest x-ray views. The advent of cross-sectional imaging with CT scan has made some of these views less common, as CT imaging is able to determine the three-dimensional location of objects with high accuracy. However, in some cases, the detailed information provided by CT is unnecessary, and more limited information from x-ray may be sufficient for clinical action such as draining a pleural effusion. Although CT could be used in this scenario, the radiation dose from CT is approximately 500 times that of a single additional chest radiograph, and the cost is approximately 10 times higher. In addition, for chest angiography protocols, chest CT requires IV contrast administration, with risks of allergy or contrast nephropathy. X-ray avoids these issues. On an individual basis, the emergency physician should consider the information needed for clinical decision-making. If it is obvious that CT will be required for specific information, such as evaluation of pulmonary embolism or aortic pathology, additional x-rays should generally not be obtained. However, if the information required can be provided by radiographs, these are a better choice for reasons of cost, radiation, and contrast exposure.

Lateral Decubitus Views

Lateral decubitus views allow assessment of radiographically visible mobile fluid collections and foreign bodies, as well as inferences about the presence of radiographically invisible foreign bodies. In a lateral decubitus view, the patient is positioned with one side of the thorax (right or left) in a dependent position (Figures 5-19 and 5-20). The view is labeled based on the side of the chest that is dependent. Thus an x-ray obtained with the patient positioned with the left side of the thorax in a dependent position is a “left lateral decubitus” view. Usually, an x-ray is then obtained using an AP projection, as described earlier.

image

Figure 5-20 Lateral decubitus x-rays and pleural effusions.

Same patient as in Figure 5-9. In this patient with a pleural effusion, a decubitus view reveals the effusion to be mobile. A right lateral decubitus view (right side down) shows fluid to layer with gravity in the right chest, emphasizing the quantity and mobility of the right pleural fluid. If the density seen on the posterior–anterior upright x-ray had been a parenchymal infiltrate, atelectasis, or a loculated pleural effusion, it would not have moved with this change in the patient’s position. Compare with Figure 5-9 and Figure 5-19, A and B.

Decubitus views can be useful in the following scenarios:

Phase of Respiration: Inspiratory and Expiratory Views

Normally, a chest x-ray is obtained at full inspiration, although in many cases patients may fail to inspire deeply because of chest pain or may be unable to hold their breath in this position because of dyspnea. The chest x-ray captures a frozen, static moment in time, but thoracic structures are actually in motion with patient respiration and cardiac activity. The apparent density of lung tissue varies with the phase of respiration. At end-expiration, lung volumes are very low, the diaphragms appear high, and lung tissue appears dense (whiter). Blood vessels in the lungs appear crowded together, contributing to the apparent density of lung tissue. If the phase of respiration is not considered, this appearance may be mistaken for pulmonary edema. In contrast, a chest x-ray obtained at end-inspiration shows well-inflated lungs, diaphragms that have descended fully, and widely spaced pulmonary vessels. Lung parenchyma under these conditions appears less dense. A clue to the phase of respiration is the number of ribs visible. A rule of thumb is that the diaphragm should lie at the level of the posterior 8th to 10th rib for an adequate respiratory effort in a good-quality chest x-ray. Abnormally dense lung parenchyma such as seen with pneumonia is more visible against the backdrop of fully inflated and lucent (black) lungs. Patients with COPD or asthma may have hyperexpanded lungs with low-density parenchyma and more than 10 ribs visible.

Occasionally, chest x-rays are intentionally obtained in other phases of respiration. Examples include:

Chest X-ray Exposure (Penetration)

The exposure of chest x-ray is particularly critical, as the chest contains structures across a wide range of densities, from air to bone. A fully exposed x-ray film or detector results in a completely black image. For example, x-ray passes readily through air outside the patient, making the background around the patient appear black. Lungs are normally composed mostly of air, with a small density contribution from pulmonary blood vessels, and appear nearly black with a good exposure. A pneumothorax is even less dense, like air outside of the patient, and appears almost completely black. Metals and bones prevent transmission of much x-ray to the detector with a normal exposure; consequently, the detector is not exposed and the image appears white.

From these principles, we can extrapolate that an overexposed chest x-ray will appear black (the entire detector is fully exposed). An underexposed chest x-ray will appear nearly white, as the detector is not exposed. Although the appearance of tissues can be adjusted (brightness and contrast) on a digital picture archiving and communication system (PACS) display, no amount of image manipulation can overcome a badly over- or underexposed image. For example, in a badly overexposed image, the entire detector is fully exposed, and all pixels are completely black. Adjusting the contrast and brightness simply makes the entire image blacker or whiter, without revealing tissue detail. In a badly underexposed image, the entire detector fails to be exposed, and all pixels are completely white. Adjustment of brightness and contrast is not useful in this case, either. As the exposure level is increased above an “optimal exposure,” lung tissue becomes “burned out” (black), and fine details of lung architecture, such as bulla, fissures, pulmonary vascularity, and pneumothorax, are lost. This loss comes with a gain in the visibility of bony detail, as x-ray can now penetrate through less dense regions of bone, including fracture zones. If exposure is lower than an “optimal level,” detail of bone is lost, as no x-ray can penetrate through dense bone. At the same time, soft tissues become more visible, as the lower exposure prevents x-ray from fully exposing the detector behind them. Depending on the clinical presentation, overexposure- or underexposure may be intentionally performed to highlight either bone or soft-tissue detail.

Beware of some common scenarios in which poor exposure can simulate pathology:

Artifacts Outside of the Patient

Opacities and lucencies on the x-ray image may not originate within the patient (Figures 5-22 through 5-25). Overlying skin folds and clothing can create lines simulating pathology such as the pleural line of pneumothorax. Foreign bodies may appear to lie within the patient but may be discerned to be outside of the patient if a lateral view is obtained. Two orthogonal views are needed to confirm an object’s location. Whenever possible, extraneous external foreign bodies should be removed before obtaining x-ray to prevent them from obscuring internal anatomy or being mistaken for objects within the patient. Sometimes surface anatomy such as nipples can simulate internal pathology such as lung nodules. When this is in doubt, marking the surface anatomy with a radiopaque tag can clarify the source of the chest x-ray finding.

image

Figure 5-23 Assessing for artifacts: Button battery ingestion.

This 7-year-old child was “washing” a button battery in her mouth in an effort to fix her electric toy when she reported accidentally swallowing the battery. Parents were uncertain whether any ingestion had occurred. When an ingested foreign body is suspected, two orthogonal views (PA or AP and lateral) are needed to localize the object. The anterior–posterior (AP) supine view shown here reveals two radiopaque objects, labeled 1 and 3 on this image. The lateral view (Figure 5-24) reveals 3 radiopaque objects, including an object not seen on this AP image but whose position is labeled 2. Did the patient ingest multiple batteries? Only object 1 is projected in a consistent location overlying the abdomen on both AP and lateral views and is intraabdominal. The other two objects are external to the patient. Object 3 on the lateral view (see Figure 5-24) corresponds to the caudad object on the AP view. Object 2 on the lateral view is so far lateral that it is not visible on the AP view. The patient underwent upper endoscopy, but the battery had moved beyond the pylorus and could not be retrieved. X-rays obtained 24 hours later confirmed passage of the battery.

image

Figure 5-24 Assessing for artifacts: Button battery ingestion.

Same patient as Figure 5-23. The lateral view shown here reveals the three radiopaque objects. (See also Figure 5-23.)

Tissue Densities and Chest X-ray

On chest x-ray, only four basic tissue densities are present (Box 5-1): air, fat, water (typical of both fluids and most solid body organs, such as liver and muscle, which are more than 90% water), and metal. This includes calcium, which you will recall from chemistry class is actually a metal. Some texts distinguish bone from metal. A key principle of chest x-ray interpretation is that two tissues in direct physical contact with each other and sharing the same basic density are indistinguishable. For this reason, the diaphragm cannot normally be discerned from the abutting liver, because both are of water density. Similarly, blood (water density) cannot be seen within the heart (also water density). Pericardial fluid cannot be distinguished from myocardium, because both are of water density, but pericardial air can be distinguished from myocardium. A pleural effusion lying above the diaphragm shares the same density with the subjacent diaphragm and liver and cannot be directly distinguished.

Normal lung tissue (air density) can be distinguished readily from adjacent normal soft tissues (water density), because the two tissues do not share the same density. However, when normally air-filled alveoli become filled with water density, the border between that lung tissue and the adjacent heart disappears. The cause of the alveolar water density does not matter: pneumonia, hemorrhage, and aspirated material all have the same effect, increasing the density of the lung from air to water. Consequently, if the abnormal lung abuts another water density structure such as the heart or diaphragm, the junction between the two tissues cannot be discerned. A little anatomy illustrates the clinical utility of this fact. The right middle lobe abuts the lower right heart border, and the right lower lobe abuts the diaphragm. For this reason, a right middle lobe pneumonia obscures the lower right heart border, and a right lower lobe pneumonia obscures the diaphragm. On the left side, the lingula abuts the lower left heart border, and the left lower lobe abuts the diaphragm. Pneumonias in these areas obscure the left heart border and left diaphragm, respectively.

Silhouette Sign

As we have just discussed, two objects with the same basic chest x-ray density (air, fat, water, or metal/bone) abutting one another cannot be distinguished. The corollary of this is that when two structures of different basic density abut each other, their junction can be discerned on x-ray. This is called the silhouette sign (Figure 5-26).27 Loss of the normal silhouette sign along the heart borders or diaphragm is an important pathologic finding, though it does not always signify pneumonia, as we explore further. Any change that causes the abnormal juxtaposition of two tissues of the same density results in loss of the silhouette sign. Examples include a solid lung mass adjacent to the heart, a pleural effusion adjacent to the heart or diaphragm, dense pulmonary edema, or atelectasis of lung adjacent to the heart or diaphragm, in which case the “collapse” of lung tissue results in it taking on a solid organ density (water density). Box 5-2 outlines the differential diagnosis of increased lung parenchymal opacity.

In other cases, the pathologic change is not the loss of the silhouette sign but rather the introduction of a silhouette sign when one is not normally present. This occurs when a new, abnormal tissue density becomes interposed between two adjacent tissues that share the same density and normally are not distinguishable. For example, the junction between the liver and the diaphragm is not normally visible, as the two tissues are in direct physical contact and share the same density (water or soft-tissue density). If pneumoperitoneum develops, air can become interposed between the liver and the diaphragm, making this interface visible (see Figure 5-11). Air can also cause a pathologic silhouette sign when it is present within the pericardium (between the soft tissues of the heart and the pericardial sac) or in the abdomen when air is visible between two adjacent loops of bowel, making the external surface of the bowel wall visible.

Air Bronchograms: A Manifestation of the Silhouette Sign

The silhouette sign can sometimes be used to distinguish otherwise similar-appearing chest x-ray findings. For example, a pleural effusion and a dense pulmonary infiltrate may have similar appearances on upright chest x-ray. Because both conditions are water density and both may abut the heart and mediastinum, both may cause loss of the normal silhouette sign. However, a pleural effusion appears uniformly dense, whereas a pulmonary alveolar infiltrate may result in the appearance of a new, abnormal silhouette called an air bronchogram (Figure 5-27). Normally, bronchi within the lung parenchyma are not visible on chest x-ray. This is because the bronchi contain air and immediately abut alveoli, which also contain air. The bronchial wall itself is too thin to attenuate x-ray to a significant degree and is therefore invisible, except when it is seen end-on. In these cases, it may be visible as a small, circular density with a hollow, lucent (black) center. This normal scenario is disrupted when alveolar fluid is present—whether that fluid is pus (pneumonia), serous fluid (pulmonary edema), blood (pulmonary hemorrhage), or lunch (aspirated food material). Under these conditions, fluid-filled alveoli become visible as water density immediately adjacent to air-filled bronchi, and the silhouette of air-filled bronchi may be seen. At extremes of the conditions described, pus, pulmonary edema fluid, blood, or aspirated material may fill bronchi as well, in which case air bronchograms may not be seen. When present, air bronchograms prove alveolar opacification. When air bronchograms are absent, either pleural fluid or alveolar and bronchial fluid may be present.

Silhouette Sign and Lateral Chest X-ray

We discussed the lateral chest x-ray earlier in some detail in the section on x-ray technique and views. However, this topic is important enough to revisit in the context of the silhouette sign.

Fissures

The major and minor fissures can provide anatomic landmarks, revealing important disease when present (Figures 5-28 through 5-30). The fissures represent pleural boundaries and are actually a “sandwich” composed of the distinct pleura lining two adjacent pulmonary lobes and the potential space between. Thickening of a fissure on chest x-ray can be the result of thickening of the pleura itself, fluid accumulating in the potential space between the two pleural layers, or disease such as malignancy. The minor fissure (also called the transverse fissure) is usually visible as a thin horizontal line on both frontal (PA or AP) and lateral chest x-ray projections. It separates the upper and middle lobes of the right lung. The major fissure is not visible on the frontal chest x-ray projection. On the lateral projection, it is visible as a thin line running from the inferior portion of the anterior thorax to the superior aspect of the posterior thorax. It is actually composed of two fissures, superimposed on the lateral projection. On the left, it divides the upper and lower lobes of the left lung. On the right, it divides the upper and middle lobes from the lower lobe of the right lung.

Fissures can change rapidly in their appearance (Figures 5-31 through 5-34). Thickening of fissures can develop or disappear in a matter of hours, sometimes giving clues to the cause. Rapidly changing fissures generally indicate fluid within the fissure, rather than pleural thickening. Changes can be dramatic; fluid in the minor fissure can assume a rounded appearance mimicking a mass lesion but disappearing within days or hours. This has been called the “vanishing tumor” or “pseudotumor.”

image

Figure 5-32 Fissures come and fissures go.

Same patient as Figure 5-31, now, with worsened heart failure. The fluid in the minor and major fissures is prominent, marking the locations of the previously invisible fissures. Also compare with Figures 5-33 and 5-34.

image

Figure 5-33 Fissures come and fissures go.

Same patient as in Figures 5-31 and 5-32, now with progressively worsening heart failure. The fluid in the minor and major fissures is prominent and rounded, with a masslike appearance. Without prior and future chest x-rays to monitor the appearance, a mass cannot be excluded. Rapidly resolving fluid in the major and minor fissures has sometimes been termed “vanishing tumor” or “pseudotumor” because of this remarkable ability to simulate a mass lesion. Compare also with Figure 5-34.

image

Figure 5-34 Fissures come and fissures go.

Same patient as in Figures 5-31 through 5-33, now after treatment for heart failure. The fluid in the minor and major fissures, which had previously assumed a rounded, masslike appearance, is nearly gone, in keeping with the term “vanishing tumor.”

Changes in Volume and Pressure on Chest X-ray

Changes in volume and pressure can be seen on frontal chest x-ray and can indicate immediately life-threatening pathology (Figure 5-35). Mediastinal shift is seen with both increased and decreased pressure and volume. Increased pressure and volume on the side with pathology push the mediastinum away from the abnormal side. Decreased pressure and volume pull the mediastinum toward the abnormal side. In both cases, tracheal deviation may be seen, in the same direction as the mediastinal shift.

Increased Pressure and Volume

Increased pressure and volume, as seen in a tension pneumothorax (Figure 5-36; see also Figure 5-35), result in the diaphragm on the abnormal side being pushed down. A tension pneumothorax on the right side may be notable for the right hemidiaphragm being lower than the left—a reversal of the normal pattern. Normally, the right hemidiaphragm is higher than the left because of the liver (right side) being larger than the spleen (left side) and displacing the diaphragm upward. In tension pneumothorax, a deep sulcus sign can be seen (see Figures 5-35 and 5-36), with the costophrenic angle on the affected side being deeper than on the unaffected side. In some extreme cases, the sulcus may be so deep that its inferior extent is not visible on the chest x-ray. Tension pneumothorax is discussed in detail in Chapter 6. We discuss pneumothoraces in more detail later, but naturally a pneumothorax is distinguished by an absence of lung markings on the affected side. Purists argue that tension pneumothorax is a clinical diagnosis, marked by hypotension and shock, and never to be found on chest x-ray. Nonetheless, chest x-rays sometimes record these findings, and an emergency physician must immediately recognize chest x-ray findings suggestive of tension pneumothorax and respond appropriately.

Decreased Pressure and Volume

Decreased pressure and volume (see Figure 5-35) may result in the diaphragm appearing higher than usual—because of the absence of aerated lung pushing it down into its normal position. In addition, the normal position of the minor fissure may be changed on the frontal x-ray. The minor fissure is usually visible as a thin horizontal line roughly at the midpoint of the cephalad–caudad diameter of the thorax. When upper lobe collapse occurs, the minor fissure is elevated, whereas lower lobe collapse may result in the minor fissure assuming a lower-than-normal position. Think of lobar collapse as resulting in an accordion-like compression of the involved lung segment, with the minor fissure following in the direction of collapse and volume loss. In addition, the collapsed lobe may be visible as an increased density. If the collapsed lobe abuts the heart, abnormal loss of the silhouette sign may occur. Common scenarios leading to lobar collapse include basilar atelectasis (Figure 5-37), bronchial mucus plugging (Figure 5-38), iatrogenic right main bronchus intubation (Figure 5-39), obstructing endobronchial lesion (e.g., carcinoma), and extrinsic compression of a bronchus from a mass lesion. Patients who have undergone pneumonectomy also have significant volume loss (Figures 5-40 and 5-41).

image

Figure 5-38 Lobar collapse and volume loss with mediastinal shift. This 70-year-old male presented with hypoxia, hypotension, and altered mental status. A, Initial chest x-ray before intubation. B, Chest x-ray after intubation was performed. The patient’s vital signs did not improve. C, Chest x-ray minutes later, after an intervention was performed and the patient’s vital signs improved. The initial chest x-ray (A) shows evidence of “volume loss.” Volume loss refers to a decrease in the size of the contents of a portion of the thoracic cavity. In contrast to tension pneumothorax, where an increase in the volume of one hemithorax displaces the mediastinum away from the affected side, volume loss means decrease in volume and causes mediastinal shift toward the affected side. Three signs point to volume loss in this patient: the mediastinum is shifted toward the patient’s right, tracheal deviation is present toward the patient’s right, and the right hemidiaphragm is elevated. In this case, the patient had volume loss resulting from lobar collapse because of mucous plugging. Intubation initially worsened the patient’s condition, because the normal (left) lung readily inflated under positive pressure. This increased the degree of mediastinal shift toward the patient’s right (B), further compressing the left lung. Adjusting the ventilator to increase positive pressure ventilation and positive end-expiratory pressure resulted in reinflation of the collapsed lobe, resolving the chest x-ray abnormalities and improving the patient’s clinical condition (C). In some patients, such as those with collapse because of an obstructing bronchial mass, this intervention might not be helpful. Bronchoscopy with stenting of the obstructed airway might be necessary to restore lobar inflation. Avoid mistaking volume loss for pleural effusion. A pleural effusion might appear similar to the elevated right hemidiaphragm, but pleural effusion occupies volume and thus should not cause mediastinal shift or tracheal deviation toward the affected side. Bedside ultrasound may help to discriminate between the two conditions. In this case, no effusion was present, and the patient’s liver was visible on ultrasound, at the level of the nipple viewed through the right axilla. Compare with Figure 5-35.

Specific Pathologic Diagnoses

So far, we have reviewed many key principles of chest x-ray interpretation, with little discussion of the specific radiographic appearance of any acute pathology. With these basic principles in mind, we now discuss key findings of common and serious emergency medical conditions. The remainder of this chapter includes nearly 200 cases of key emergency diagnoses, with case descriptions and annotated figures displaying radiographic findings. To guide you through these figures, we have summarized the content of figures (Table 5-5).

TABLE 5-5 Arrangement of Figures

Figure content Figure number
acute chest syndrome 91
air bronchograms 27, 53-54
aortic pathology see Chapter 7
aspiration pneumonitis 88-90
asthma 42
atelectasis 37
cardiac CT see Chapter 8
cardiomegaly 104-115, 131-135
cavitary lesions/tuberculosis 92-96
central venous catheter 188
changes in pressure and volume 35
chest x-ray technique 1,2,3,4, 5, 6,7
congestive heart failure/pulmonary edema 116-122
chronicity of chest x-ray findings 52
chronic obstructive pulmonary disease 43-48
cystic fibrosis 49
dextrocardia 198
esophageal foreign body 187
esophageal rupture 157
external objects 22, 25
fissures 28, 29-34
hiatal hernia 185-6
hilar adenopathy 183-184
idiopathic pulmonary fibrosis 50
internal foreign bodies 23, 24
lateral decubitus x-ray 19, 20
lobar collapse 38, 39
lung abscess 97-101
lung masses 159-175
mediastinal widening 176-178
meniscus sign 8
nasogastric tube 189
normal chest x-ray 15, 16
pacemaker complications 190-197
pericardial effusion 123-130
pertussis 86-87
pleural effusions 9-10, 136-152
pneumonectomy 40, 41
pneumonia 55-85
pneumopericardium 153-156
pneumoperitoneum 11, 12, 13, 14
pulmonary embolism see Chapter 7
retrosternal space 17
right main bronchus intubation 39
rotation artifact on chest x-ray 21
septic emboli 102-103
silhouette sign 26
smoke inhalation 51
spine sign 18
subcutaneous emphysema 158
superior vena cava syndrome 179-182
tension pneumothorax 36
traumatic injuries see Chapter 6

Findings of Common and Serious Emergency Pathology

Asthma and Chronic Obstructive Pulmonary Disease

Both asthma and COPD (Figures 5-42 and 5-43) are disease processes characterized by air trapping and lung hyperinflation. Consequently, they often share the findings of flattened diaphragms (because of hyperinflated lungs pushing the diaphragms down) and increased lung volumes, AP chest diameter, and retrosternal space. The lateral view is useful for assessment of the latter findings and may show the diaphragms to be flattened to a greater extent than was visible on the frontal projection. On the frontal view, the heart may appear small because of the hyperinflated lung fields. In patients with COPD, bullae (also called blebs) may develop (Figures 5-44 through 5-48). These regions are devoid of lung markings and must be carefully distinguished from pneumothorax, as large peripheral bullae can have a similar appearance. Bullae can displace normal pulmonary blood vessels, sometimes crowding these vessels together. Patients with COPD and asthma can suffer barotrauma—spontaneously or because of mechanical ventilation. The chest x-ray should be inspected carefully for pneumothorax and pneumopericardium.

image

Figure 5-43 Chronic obstructive pulmonary disease (COPD).

A, Posterior-anterior (PA) upright chest x-ray. B, Lateral upright chest x-ray. This 63-year-old male with a history of COPD presented with 2 weeks of worsening cough with yellow sputum and dyspnea. His oxygen saturation was 87% in the emergency department. As in Figure 5-42, this patient has evidence of hyperinflation, with flat diaphragms (particularly evident on the lateral x-ray, B). The patient also has a blunted right costophrenic angle with an apparent effusion and increased densities in both the right and the left lung base (A). The lateral x-ray also shows increased density overlying the inferior thoracic spine, an abnormal spine sign. Findings of pleural effusions and pneumonia are discussed in detail in other figures.

image

Figure 5-45 Bullae in chronic obstructive pulmonary disease.

Close-up of bullous changes from the same patient as Figure 5-44. The walls of the large bullae are visible, and a paucity of lung markings makes these regions appear abnormally lucent (black).

image

Figure 5-47 Bullae in chronic obstructive pulmonary disease (COPD).

Same patient as Figure 5-46. In the lung apex, the normal lung has been replaced by large bullae, which are air-containing regions lacking the normal parenchymal supporting structures and blood vessels. Consequently, these areas do not attenuate the x-ray beam and look blacker than normal lung on chest x-ray. Compare with the CT scan from the same patient in Figure 5-48.

image

Figure 5-48 Bullae in chronic obstructive pulmonary disease (COPD).

CT scan from the same patient as Figures 5-46 and 5-47, viewed on lung windows. Slice at the level of the aortic arch (A), through the pulmonary artery (B and C), and through the level of the left atrium (D). Note the severe bullous changes in the more cephalad slices. D, The lung parenchyma appears nearly normal in this more caudad slice.

Other Forms of Chronic Lung Disease

Chronic lung diseases such as cystic fibrosis (Figure 5-49), idiopathic pulmonary fibrosis (Figure 5-50), and chronic injuries from smoke inhalation (Figure 5-51) have a similar appearance on chest x-ray. Dense interstitial markings may be seen, and air space disease may be chronically present. Bullae such as those seen in COPD may be present. Often, the appearance is quite abnormal, and comparison with prior chest x-rays is essential to distinguish new opacities suggesting acute infection from chronic disease. These diseases serve as an important reminder that chest x-ray findings should not be assumed to be acute and to explain new symptoms. Comparison with prior x-rays should always be performed when possible (Figure 5-52).

Pneumonia

Pneumonia can follow several common patterns on chest x-ray. First, air space opacification may be present, if alveolar fluid and cellular debris develops. This pattern is common in patients with “typical” bacterial infections such as Streptococcus pneumoniae. Air bronchograms (Figures 5-53 and 5-54; see also Figure 5-27) may be present, as described in earlier sections. Often, the increased density of lung parenchyma obscures the margins of adjacent solid organs such as the heart or diaphragm, resulting in a pathologic loss of the silhouette sign as described earlier. On a frontal projection, a right middle lobe pneumonia obscures the lower right heart border, and a right lower lobe pneumonia obscures the right diaphragm. A left lower lobe pneumonia obscures the left diaphragm, and an infiltrate in the lingula obscures the left heart border (remember that the left lung does not have a middle lobe). Upper lobe pneumonias may obscure the borders of the upper heart and mediastinum. On the lateral view, a pathologic spine sign (described earlier) may be seen, with increased density in the retrocardiac space and overlying the lower thoracic spine. A pleural effusion may be present with pneumonia. If infection is confined to a lobe, the increased density may extend to, but not beyond, a fissure (described earlier) on the frontal or lateral views. Volume loss (described earlier) may occur if mucous plugging is present, with resulting findings including deviation of fissures, elevation of the diaphragm, or mediastinal and tracheal shift. Figures 5-55 through 5-85 depict pneumonias with air space disease in patients of varying ages, and with varying lung regions involved. Pneumonia remains a key diagnosis because of its mortality in patients at extremes of age. The appearance can be highly variable, and chest x-ray is not always diagnostic. Review the figures to gain perspective on the potential appearance of infiltrates.

image

Figure 5-59 Pneumonia: Lingula and left lower lobe.

A, Posterior-anterior (PA) upright chest x-ray. B, Lateral upright chest x-ray. Same patient as Figure 5-58, now 5 days later. The patient returned with worsened symptoms. A, Note how the lingular infiltrate has now significantly hidden the left heart border. B, The retrocardiac left lower lobe infiltrate persists.

image

Figure 5-64 Pneumonia: Right middle lobe and left upper lobe.

A, Anterior-posterior (AP) chest x-ray. B, Close-up from A. Same patient as Figure 5-63, where his initial chest x-ray showed a classic right middle lobe pneumonia. This x-ray was obtained one week later, with a progressive density appearing to involve the entire right middle lobe. A, The right heart border remains obscured, and the right diaphragm is also partially hidden, which may indicate right lower lobe involvement or an associated pleural effusion. Increased density is also seen in the bilateral upper lungs. B, An air bronchogram is faintly visible behind the left heart.

image

Figure 5-67 Pneumonia: Right upper lobe on computed tomography (CT) with IV contrast.

Same patient as Figure 5-66. CT images show the extent of parenchymal opacification. Lung windows are shown; the soft-tissue windows were negative for pulmonary embolism. A, Axial CT lung window image. B and C, Coronal CT lung window images.

image

Figure 5-69 Pneumonia: Bilateral (Streptococcus pneumoniae) on computed tomography (CT) lung windows.

Same patient as Figure 5-68. On lung windows, dense infiltrates are seen in the right posterior lung and lingula. Note how, even on CT, a lingular infiltrate hides the left heart border. A clear air bronchogram with a treelike form is seen because of adjacent alveoli filled with fluid. See how the lingular infiltrate is tightly circumscribed along its posterior border by a pleural boundary.

image

Figure 5-73 Pneumonia: Left upper lobe.

Same patient as in Figure 5-72. A, Left lateral decubitus x-ray. B, Right lateral decubitus x-ray. Decubitus views were obtained in an attempt to identify an effusion if present. No significant change is seen in the density of the left lung with these positional changes. The left costophrenic angle appears slightly more lucent when the left chest is positioned higher, suggesting that a small effusion may have contributed to its density when the left chest is down. Computed tomography was performed to further evaluate this (Figure 5-74).

image

Figure 5-79 Pneumonia: Right lower lobe on CT lung windows.

Same patient as Figure 5-78. CT was performed because of concern for abscess, as the patient had abdominal pain and an indwelling dialysis catheter that could be a source for bacteremia. On lung windows, an area of increased density is visible in the right lower lobe, corresponding to the area of increased density on her lateral chest x-ray (Figure 5-78). This is likely pneumonia—the patient’s blood cultures grew Staphylococcus aureus. A, Axial view. B, Coronal view.

image

Figure 5-85 Pneumonia: Pediatric, right upper lobe.

A, Frontal projection chest x-ray. B, Lateral chest x-ray. Same patient as Figure 5-84. Follow-up x-rays the next day showed increased density along the right heart border, suggesting progression of pneumonia. Note the increased upper chest density on the lateral x-ray as well. On the lateral view, the retrosternal space appears dense because of the presence of the normal thymus.

A second common pattern of pneumonia is an interstitial pattern, often seen with “atypical” pathogens such as Bordetella pertussis (Figure 5-86), Chlamydia pneumonia, and Mycoplasma. In this circumstance, fluid becomes evident in the interstitial connective tissues of the lung, rather than within alveoli. Air bronchograms should not be seen as a result of an interstitial pneumonia. Peribronchial cuffing may be seen. Peribronchial cuffing (Figure 5-87) is a thickened appearance of bronchioles when viewed end-on on chest x-ray. Normal bronchioles are not visible when oriented in the plane of the x-ray but are visible as small black circles (air within the bronchiole) surrounded by a thin white line (the bronchial wall) when they are oriented perpendicular to the plane of the chest x-ray. With the development of interstitial edema, this thin white line thickens and is then called peribronchial cuffing. Peribronchial cuffing is not specific to bacterial infection and is seen in other disease states, including pulmonary edema, asthma, pneumocystis pneumonia (PCP), viral pneumonitis, and bronchiolitis such as that from respiratory syncitial virus. Other findings of interstitial pneumonia can include fluid in the minor and major fissures.

image

Figure 5-87 Pertussis and peribronchial cuffing.

Close-up from Figure 5-86, B. What is peribronchial cuffing? Peribronchial cuffing is edema of bronchioles, resulting from diverse causes including asthma, viral illnesses, heart failure, and pertussis. The bronchial wall becomes thickened and denser from edema fluid. Like an air bronchogram, this creates contrast with the air-filled bronchiole lumen, making it more visible. Normally, bronchioles are visible only when they are oriented perpendicular to the image plane. They are thin-walled and appear as discrete circles in short-axis cross section. When edema is present, the bronchiolar wall is thickened in cross section, and bronchioles may be seen in other orientations, such as long-axis cross section, because of their increased density. Look at the close-up of the hilum, taken from the lateral x-ray in the prior figure. You will see one bronchiole in short-axis cross section and several others in long-axis cross section. The density surrounding the air-filled lumen is the edematous bronchial wall.

Pneumonia Differential Diagnosis

It is essential to remember that a differential diagnosis exists for increased lung parenchymal density (see Box 5-2). Any process that results in fluid and solid material accumulating within alveoli can have this x-ray appearance. This includes not only infectious pneumonitis (pneumonia) but also physical aspiration of noninfectious material (Figures 5-88 through 5-90), pulmonary hemorrhage, pulmonary infarction in the setting of pulmonary embolism (see Chapter 7) or sickle cell anemia acute chest syndrome (Figure 5-91), atelectasis, mucous plugging with segmental collapse, mass lesion, and pulmonary edema. A pleural effusion can resemble parenchymal opacity, with some differences as outlined in earlier sections. When a parenchymal opacity is seen, the emergency physician must resist the temptation to conclude that “the chest x-ray shows pneumonia.” Instead, the physician must determine that “the chest x-ray shows increased parenchymal opacity.” This finding should be considered with other clinical data to reach the correct diagnosis. In the words of a radiologist, clinical correlation is required.

image

Figure 5-90 Aspiration: Gasoline.

Same patient as Figure 5-89. In a chest x-ray obtained 24 hours after presentation, his right middle lobe infiltrate has progressed. A, Posterior-anterior (PA) chest x-ray. B, Lateral chest x-ray.

image

Figure 5-91 Acute chest syndrome: Sickle cell anemia.

Acute chest syndrome of sickle cell anemia does not have a single pathognomonic appearance on chest x-ray. The condition is caused by microvascular sludging of sickled red blood cells, resulting in infarction of the lung from small vessel occlusion. In the face of parenchymal necrosis, fluid accumulates within alveoli, increasing the density of lung tissue from air density to soft-tissue density (fluid density). The radiographic appearance is indistinguishable from that of other pneumonias. Because the infarction occurs in small vessels, the distribution of infarction is not in the territory of a large vessel as might be seen with pulmonary embolism, where infarction of an entire lung segment distal to the point of thrombotic occlusion may be seen. Thus the distribution of lung infiltrates in acute chest syndrome may be patchy or may follow a lobar distribution. Because alveoli fill with fluid, air bronchograms may be seen, as in bacterial pneumonia. Making the diagnosis even more indistinct, patients with sickle cell anemia may have bacterial pneumonia with negative blood and sputum cultures, so a pulmonary infection may be wrongly labeled as acute chest syndrome. With these caveats, this x-ray is from a 24-year-old male with sickle cell anemia disease, presenting with fever and an oxygen saturation of 60% on room air. He improved with oxygen, fluids, and antibiotic administration, although antigen testing for Streptococcus pneumoniae, Legionella, and influenza were negative, as were cultures. A, The posterior-anterior (PA) chest x-ray demonstrates patchy bilateral infiltrates. The right lower lobe shows particular density, and increased density is seen in the retrocardiac space on the lateral x-ray (B)—note the absence of the usual decrease in density toward the diaphragm (see text introduction to chest x-ray and text section on the spine sign).

Cavitary Lesions

Some pulmonary infections characteristically produce cavitary lesions on chest x-ray. These include tuberculosis and lung abscesses from pathogens such as Staphylococcus aureus or aspirated anaerobes. Necrotic neoplastic lung masses may cavitate. Bullae resulting from emphysema can become secondarily infected, sometimes with fungal pathogens. A typical cavitary lesion may have an air–fluid level on an upright chest x-ray. This may not be visible on a supine image, as fluid within the cavity layers in the same plane as the x-ray detector. As with other chest x-ray findings, a differential diagnosis should be considered based on appearance and other clinical data. Figures 5-92 through 5-101 show cavitary lesions worrisome for tuberculosis or abscess.

image

Figure 5-93 Tuberculosis or cavitary lesion.

Chest CT without IV contrast from the same patient as Figure 5-92. Soft-tissue (A) and lung (B) windows show a well-circumscribed rounded lesion 4 cm in diameter with an air–fluid level. Given his immunosuppressed status, the differential diagnosis included tuberculosis, lung abscess, neoplasm, and fungal infection. He was readmitted for video-assisted thorascopic surgery and left upper lobe wedge resection. Pathology revealed only an organizing hematoma.

image

Figure 5-96 Tuberculosis or cavitary lesion.

Same patient as Figure 5-95. Chest CT without IV contrast shows dense consolidation of the right upper lobe, with air bronchograms and multiple cavitary regions with air–fluid levels. These may represent infected bullae, areas of necrotizing pneumonia and abscess formation, or tuberculous cavitation. CT cannot discriminate among these causes. Left column (A, C, E, G, I), axial slices viewed on lung windows, progressing from cephalad to caudad. Right column (B, D, F, H, J), axial slices at the same anatomic levels as A, C, E, G, and I viewed on soft tissue windows.

image

Figure 5-98 Lung abscess.

Same patient as Figure 5-97. Chest CT with intravenous contrast shows a lung abscess located immediately right of the thoracic spine. The thick rim enhances with contrast, giving it a brighter white appearance. The center of the abscess contains fluid density (dark gray) with a tiny focus of air. This abscess corresponds to the retrocardiac density seen on the patient’s chest x-ray. No air–fluid level was seen on chest x-ray because the abscess contains only small gas bubbles. A, Axial slice viewed on soft-tissue windows. B, The same slice viewed on lung windows. C, Close-up from A.

image

Figure 5-100 Lung abscess.

Same patient as Figure 5-99. Chest CT without IV contrast. A, Lung windows. B, Same slice viewed on soft-tissue windows. This 9-cm left lung abscess has a thick rind and an air–fluid level. The patient was treated with antibiotics and 2 weeks later began to cough up grossly purulent sputum. Repeat CT showed a decreased air–fluid level suggesting spontaneous drainage of the abscess (Figure 5-101).

image

Figure 5-101 Lung abscess.

Same patient as Figures 5-99 and 5-100. Chest CT without IV contrast. A, Lung windows, B, Soft-tissue windows. This chest CT was performed after 2 weeks of antibiotic therapy and showed a decreased air–fluid level suggesting spontaneous drainage of the abscess (compare with prior figure).

Embolic Pneumonia

In the case of endocarditis, septic emboli may be distributed throughout the lungs. The typical appearance is of numerous or innumerable nodules, often bilaterally distributed (Figures 5-102 and 5-103). The chest x-ray appearance is indistinguishable from metastatic disease.

image

Figure 5-103 Septic emboli.

Chest CT with IV contrast from the same patient as Figure 5-102, viewed on lung windows. A, B, Adjacent axial slices at the level of the proximal aorta and pulmonary artery. C, A more caudad slice through the heart. Multiple peripheral nodules are seen in the slices shown, labeled with small arrows. This is consistent with septic emboli, although metastatic disease could have the same appearance and cannot be differentiated on CT. However, the patient had no known primary malignancy, and his clinic course suggests that these were indeed septic emboli. The patient’s blood cultures became positive for coagulase-negative Staphylococcus epidermidis and Enterococcus faecalis. He subsequently developed a septic ankle requiring washout in the operating room. His echocardiogram showed no valvular vegetations, and his dialysis catheter was removed, as it was the suspected source of infection.

Pulmonary Edema and Heart Failure

Chest x-ray findings of pulmonary edema follow a sequential pattern from early or mild findings to late or advanced findings (Table 5-6). A number of artifacts can simulate these findings and should be recognized to avoid misdiagnosis (Table 5-7). In general, early findings include pulmonary venous congestion, which is associated with cephalization, and cardiomegaly. Cephalization (Figures 5-104 and 5-105) refers to the redistribution of visible pulmonary vessels into the upper lung fields as pulmonary vascular volume and pressures increase. On a normal upright chest x-ray, the upper lung fields are relatively sparsely populated with visible pulmonary vessels compared to the lung bases because of the effect of gravity in preferentially filling dependent vascular beds. Remember that cephalization can be simulated by a supine chest x-ray, in which the upper pulmonary vessels are filled because of the patient’s supine position. An underpenetrated (underexposed) chest x-ray may simulate this appearance as well, as all blood vessels and soft tissues appear more prominent. A poor inspiratory effort also increases the apparent density of lung tissue, simulating pulmonary edema. In a poorly expanded lung (because of poor inspiratory effort at the moment of the x-ray), pulmonary blood vessels appear more closely spaced and may be mistaken for edema.

TABLE 5-6 Chest X-ray Findings of Pulmonary Edema

Finding Description
Cephalization Pulmonary vascular markings are redistributed to upper lung zones on an upright x-ray.
Cardiomegaly The ratio of the heart to the greatest transverse rib cage has a diameter >0.5.
Pericardial effusion This may simulate cardiomegaly, but a water-bottle appearance is classically present.
Interstitial edema Fluid appears in interstitial regions of the lung.
Thickening of the walls of bronchioles is seen when viewed end-on, perpendicular to the chest x-ray plane.
Parallel fine lines extend from the pleural surface into the subpleural lung, especially in lung bases, thought to result from thickened interlobular septa.
Alveolar fluid with or without air bronchograms Air bronchograms prove the presence of alveolar fluid. Their absence does not disprove the presence of alveolar fluid. Air bronchograms do not result from pleural effusion alone.
Pleural effusion, including fluid within lung fissures Right-sided pleural fluid is more common than left when a unilateral effusion occurs from heart failure.

TABLE 5-7 Chest X-ray Techniques and Artifacts That May Simulate Pulmonary Edema

Technique or Artifact Effect
Supine technique Mimics cephalization and cardiomegaly
AP technique Mimics cardiomegaly
Poor inspiratory effort or hypoinflated lungs Mimics interstitial edema and prominent pulmonary vascularity
Underexposure, e.g., in the obese patient Mimics interstitial edema and pulmonary vascular congestion
image

Figure 5-105 Cardiomegaly with cephalization.

Same patient as Figure 5-104. A, Posterior-anterior (PA) chest x-ray. B, Lateral chest x-ray. The patient returned to the emergency department for similar symptoms 13 days after his initial visit. He was diagnosed by the emergency physician with bronchitis and discharged. The radiology report notes stable cardiomegaly and mild pulmonary edema. Is the heart size really unchanged? Notice how the pulmonary vascular markings are prominent in the upper lung fields. This redistribution of vascular markings is called cephalization and is an early finding of pulmonary edema.

Cardiomegaly (Figures 5-106 through 5-109; see also Figures 5-104 and 5-105) is another frequent finding in cases of pulmonary edema, though not all patients with heart failure will have an enlarged cardiac silhouette. Heart size is measured on chest x-ray using the cardiothoracic ratio, which equals the ratio of the transverse heart to the greatest internal diameter of the thoracic cage. A normal value is less than or equal to 0.5 for adults, and cardiomegaly is radiographically defined by values exceeding this. Poor inspiration, supine positioning, or magnification from an AP portable x-ray all may falsely elevate the cardiothoracic ratio, simulating cardiomegaly. Figures 5-110 through 5-122 show cardiomegaly, often in association with other findings of pulmonary edema.

image

Figure 5-106 Cardiomegaly.

Same patient as Figures 5-104 and 5-105. A, Posterior-anterior (PA) chest x-ray. B, Lateral chest x-ray. The patient returned again 7 days later (20 days after his initial presentation). He continued to complain of cough, and his vital signs were notable for tachycardia to 130 and a temperature of 38.3°C. An interval increase in heart size was noted by the emergency physician, and a viral cardiomyopathy was suspected. The x-ray shows increased cardiomegaly with some increased vascular markings, suggesting mild pulmonary edema with cephvalization.

image

Figure 5-107 Cardiomegaly.

Two days after Figure 5-106, the patient is in the cardiac intensive care unit. The heart appears slightly larger on this view, although the supine anterior–posterior technique may exaggerate any difference. Notably, the patient has a right internal jugular catheter that curls into the right pulmonary artery—a Swann-Ganz catheter for measurement of pulmonary capillary wedge pressures. In addition, a radiopaque marker below the aortic knob marks an intra-aortic balloon pump.

image

Figure 5-108 Cardiomegaly.

Same patient as Figure 5-107, 6 days later. A, Frontal projection chest x-ray. B, Lateral chest x-ray. The patient now has a left ventricular assist device implanted in the abdomen. This device pumps blood from the left ventricle to the ascending aorta. Bilateral pleural effusions are present, masking the diaphragms. Infiltrates representing edema are also present. Remarkably, the patient received a heart transplant 6 weeks later and did well. Review Figure 5-109.

image

Figure 5-109 Cardiomegaly progression.

This series illustrates the progression of cardiomegaly in the same patient as Figures 5-104 through 5-108 over a period of 30 days. Notice the increase in heart size over this period, the increased cephalization of vascular markings, and the development of bilateral effusions and frank pulmonary edema as you move from A to E. When interpreting a chest x-ray, always compare with prior x-rays if available. A finding on the most recent x-ray may turn out to be old and thus unlikely to be the cause of acute symptoms. Or a subtle finding on the current x-ray may represent a significant change from prior x-rays and thus be important.

image

Figure 5-119 Pulmonary edema.

Same patient as in Figure 5-118. A, Frontal projection chest x-ray. B, Close-up from A. Pulmonary edema is an important finding to recognize, so take a moment with this x-ray. Note the hazy upper lung fields, the dense lower lungs with prominent interstitial markings, and the indistinct hemidiaphragms and heart borders. B, Peribronchial cuffing is also present.

image

Figure 5-120 Pulmonary edema, resolved.

Same patient as Figure 5-119, 1 month later. Note the clear diaphragms and heart borders, the decrease in vascular markings in the upper lungs, and the resolution of lower lung interstitial markings and peribronchial cuffing.

An enlarged cardiothoracic ratio may also occur with pericardial effusion (Figures 5-123 through 5-130). This can be indistinguishable from cardiomegaly on chest x-ray, though classically the appearance of a large pericardial effusion is described as a “water bottle” appearance on upright chest x-ray. Imagine the appearance of a full water balloon that is placed on a flat surface. The balloon sags, creating a broader inferior aspect. A large pericardial effusion may have a similar appearance when the heart and pericardium rest in an upright position on the diaphragm and solid liver beneath. Although chest x-ray may assist in this diagnosis, ultrasound (see Chapter 6) and CT scan are definitive and should be used to differentiate cardiomegaly from pericardial effusion when clinical concern exists. Figures 5-131 through 5-135 (see also Figures 5-123 through 5-130) show the similar chest x-ray appearance of cardiomegaly and pericardial effusion. Pericardial effusion may accompany cardiomegaly in severe heart failure. Diagnostic imaging does not definitively determine the cause of a pericardial effusion, although associated findings such as lung opacities consistent with metastatic disease or primary lung neoplasms may suggest the cause.

image

Figure 5-124 Massive pericardial effusion and tamponade.

Same patient as Figure 5-123. His preoperative (A) and postoperative (B) chest x-rays are shown from his aortic valve replacement. These preceded his presentation in Figure 5-123.

image

Figure 5-125 Massive pericardial effusion.

Same patient as Figures 5-123 and 5-124. The patient underwent chest computed tomography (CT) without (A) and then with (B) intravenous contrast to evaluate his aorta, which was normal. However, the CT confirmed a massive pericardial effusion surrounding a normal-appearing heart. Without contrast, note that fluid blood within the chambers of the heart has a slightly lower density than the pericardial effusion, which has a density more similar to that of myocardium. When contrast is administered, the ventricular chambers fill completely, and the myocardium enhances and becomes somewhat brighter than the surrounding pericardial effusion. The heart itself is outlined by a thin stripe of fat, which appears nearly black on soft-tissue windows. The patient developed hypotension, suggesting cardiac tamponade, and pericardial window was performed for drainage of the effusion. In the operating room, the effusion was found to be coagulated blood.

image

Figure 5-127 Pericardial effusion and tamponade (patient underwent pericardial window).

This 58-year-old lung transplant patient presented with 1 day of gradual dyspnea and cough. The patient had recently been admitted with Pseudomonas bacteremia and pulmonary embolism, was treated simultaneously with ciprofloxacin and coumadin, and had an INR of 2.5. Initial chest x-ray (A) was read as normal. Chest computed tomography (Figure 5-128) was negative for pulmonary embolism but showed a moderate pericardial effusion. The patient was admitted but then developed increased pericardial fluid (B) with pericardial tamponade requiring pericardial window. Remember that pericardial tamponade does not require a large volume pericardial effusion. Chest x-ray can be relatively insensitive for pericardial fluid. Consider bedside echocardiogram to assess for pericardial fluid in patients with dyspnea, chest pain, or hypotension.

image

Figure 5-128 Pericardial effusion and tamponade.

Same patient as Figure 5-127. The patient’s chest CT with IV contrast shows a moderate pericardial effusion. On soft-tissue windows, unenhancing fluid such as pericardial or pleural effusions appears as an intermediate gray. The pericardial effusion surrounds a nearly black band, which is pericardial fat (fat is nearly black on soft-tissue windows). Within this lies the heart itself—the myocardium enhances with contrast and appears somewhat brighter than the effusion. A, B, Axial slices through the heart viewed on soft-tissue windows. C, Close-up from A.

image

Figure 5-130 Pericardial effusion and tamponade.

Same patient as Figure 5-129. Chest CT with IV contrast. The patient has a large pericardial effusion. A, Axial CT slice just below the aortic arch, viewed on soft-tissue windows. B, A more caudad slice.

image

Figure 5-132 Cardiomegaly or pericardial effusion?

Computed tomography scout image from the same patient as Figures 5-131 and 5-133 again shows an enlarged cardiac silhouette. It is not clear from this view whether cardiomegaly or pericardial effusion is at play.

image

Figure 5-133 Cardiomegaly.

Chest CT from the same patient as Figures 5-131 and 5-132, without (A) and with (B) contrast. In these images, the heart appears significantly enlarged without a pericardial effusion. Without contrast, blood within the chambers of the heart are difficult to distinguish from myocardium. When contrast is injected, the chambers become an intense white, whereas the myocardium enhances somewhat.

image

Figure 5-134 Pericardial effusion and cardiomegaly.

Same patient as in Figure 5-133. Here, a small pericardial effusion is visible in a dependent position deep to the left ventricle. The ventricle appears quite dilated as well. The pericardial fluid is dark gray and does not enhance with intravenous contrast (compare A, without contrast, and B, with contrast). The myocardial wall does enhance and becomes visible in B.

image

Figure 5-135 Pericardial effusion and cardiomegaly.

Same patient as in Figure 5-134. Lower in the chest, the pericardial effusion appears larger. The pericardial fluid is dark gray and does not enhance with intravenous contrast (compare A, without contrast, and B, with contrast). The myocardial wall does enhance and becomes visible in B.

Later in the course of heart failure or developing pulmonary edema, increased hydrostatic pressure results in the development of interstitial edema. This is manifested on chest x-ray as peribronchial cuffing (see Figure 5-119) described earlier in the section on interstitial pneumonia and Kerley B lines. Peribronchial cuffing is not specific to pulmonary edema and is seen in other states, including viral pneumonitis and bronchiolitis, such as that from respiratory syncitial virus. Kerley B lines are parallel, fine lines extending from the pleural surface into the subpleural lung, especially in lung bases, and thought to result from fluid accumulation in interlobular septa.

Late in the course of pulmonary edema, hydrostatic pressure forces fluid into alveoli, usually first in the lung bases and then in the upper lung segments (see Figure 5-110, B). This is usually symmetrical, although asymmetrical pulmonary edema can occur. Before frank air space edema occurs, symmetrical perihilar “butterfly” or “bat wing” air space disease may be seen. As described earlier in the section on air bronchograms, fluid within alveoli arising from heart failure may create a silhouette against the adjacent air-filled bronchi—an air bronchogram. The presence of this finding proves the presence of alveolar fluid but does not distinguish edema from infection, hemorrhage, aspiration, or other fluid. If edema progresses further, fluid may fill bronchioles adjacent to alveoli, eliminating the silhouette of the air bronchogram.

Pleural effusions are another common finding in pulmonary edema, although they may occur from many other causes, including infection, malignancy, and pulmonary embolism. When a unilateral pleural effusion occurs in the setting of heart failure, a right-sided pleural effusion is more common. Figures 5-136 through 5-152 show a variety of pleural effusions, emphasizing their variable size and appearance by chest x-ray, ultrasound, and CT.

image

Figure 5-138 Pleural effusions.

Same patient as in Figure 5-137. The patient underwent decubitus x-rays to further delineate her pleural effusions. A, Left lateral decubitus chest x-ray. B, Right lateral decubitus chest x-ray. Decubitus views are named for the side positioned down. An x-ray obtained with the right side down is a right decubitus view. In the case of this patient, the left pleural effusion is seen to layer along the lateral chest wall in the left decubitus view—compare with the upright view in Figure 5-137, which shows the upper lateral left chest to be clear. On the right decubitus view, fluid fills the minor fissure, indicating that fluid is also mobile in the right pleural space. A meniscus remains visible in the left thorax on the right decubitus view, suggesting partial loculation of the left pleural effusion. Note that the lung in the nondependent position looks more lucent in these examples as fluid drains to a dependent position. The nondependent lung also is better aerated, with less atelectasis and less overlying soft tissue.

image

Figure 5-140 Pleural effusions.

Same patient as in Figure 5-139. An ultrasound of the chest was performed in preparation for thoracentesis. The large black region is pleural fluid. Lung tissue is seen deep to this and has echo characteristics similar to abdominal solid organs because it is atelectatic because of the pleural effusion. Thoracentesis should be possible here with little risk of pneumothorax, because 10 cm of pleural fluid separates the chest wall and lung in this location.

image

Figure 5-149 Pleural effusions.

Same patient as Figure 5-148. A, CT with IV contrast, axial slice viewed on soft-tissue windows. B, Same slice viewed with lung windows. A moderate left pleural effusion is seen. Computed tomography (CT) cannot discriminate the cause of a pleural effusion, other than by excluding parenchymal masses and pulmonary emboli. The density of pleural fluid can be measured on CT, but this is relatively nonspecific, as blood, serous fluid, and infected pleural fluid can have overlapping densities. Aspiration of pleural fluid for cytology, cultures, and chemical testing is necessary to further delineate the cause. No pulmonary emboli or masses were found on the patient’s CT.

image

Figure 5-152 Pleural effusions.

Same patient as Figure 5-151. CT with IV contrast was negative for pulmonary embolism but showed bilateral pleural effusions, including a loculated right anterior effusion that did not move with gravity. A, Lung windows. B, Same slice viewed on soft-tissue settings. On lung windows, the dependent lung appears denser—this is likely due to compressive atelectasis from the adjacent pleural effusions.