Imaging of the Thoracic Surgery Patient



General Considerations

  • A variety of imaging options are available to thoracic surgeons for evaluation of the thorax.

  • The choice of imaging examination should be based on the clinical question, recommendations in the literature, and expertise of the radiologist.

  • Cost-effective practice requires selection of procedures that increase diagnostic yield and minimize costs. (Common indications for thoracic computed tomography [CT, Table 1-1 ]. )

    TABLE 1-1 ▪


    Determination of presence and extent of neoplastic disease
    Diagnosis of pulmonary embolism
    Diagnosis of bronchiectasis, diffuse lung disease, emphysema and small airways disease.
    Guidance for interventional procedures
    Localization of loculated collections of fluid if ultrasound is not diagnostic
    Evaluation of suspected mediastinal abnormalities seen on chest x-ray examination

Chest Radiographs

  • Plain chest radiographs ( Fig. 1-1 ) are the most commonly performed radiologic investigation for the diagnosis of thoracic disease. The standard projection is an erect posteroanterior (PA) chest radiograph taken in full inspiration. The patient stands with the front of their chest against the imaging plate and the x-ray beam passes from posterior to anterior, limiting magnification of the heart and mediastinal structures. A lateral examination is useful in identifying abnormalities in regions not clearly seen on the frontal view.

    Figure 1-1

    A, Normal PA and B, lateral chest radiographs: anatomic features.

  • Patients who are unable to have a PA chest radiograph may have a portable examination at the bedside. The imaging plate is positioned against the patient’s back, with the x-ray beam passing from anterior to posterior. The beam diverges around the heart and mediastinum, resulting in magnification of these structures and often creating difficulties with interpretation. Images are made with lower energy x-rays and exposure times are longer. Ill patients often have difficulty with breath holding and the resultant respiratory motion artifact may degrade image quality.

  • Lateral decubitus views may help in the diagnosis of a pleural effusion and expiratory views may aid diagnosis of a small pneumothorax.

  • Traditional film-screen methods are being widely replaced by digital imaging. Digital systems have several advantages over film-screen systems. Images may be electronically manipulated such that an interpretable image can be generated even if the x-ray exposure is suboptimal. Images can be transmitted and stored electronically, and utilizing a picture archiving and communications system (PACS), digital images are available to referring clinicians and radiologists simultaneously both on and potentially off the hospital site.

Computed Tomography

  • CT scanners ( Figs. 1-2 and 1-3 ) generate a cross-sectional image of the patient with a very narrow beam of x-rays. This beam passes through the patient to strike electronic detectors. During the scan, the beam rotates around the patient and multiple measurements are made. Data are processed by computers, and digital images are reconstructed, free from superimposition of overlying structures.

    Figure 1-2

    Computed tomography (CT) thorax soft tissue windows demonstrating normal anatomy. A, At the level of the great vessels, the brachiocephalic vein is crossing the mediastinum from left to right anterior to the innominate, carotid, and subclavian arteries. The esophagus can be identified posterior to the trachea (T). B, At the level of the aortic arch (AA), the azygos vein can be seen lateral to the trachea (T) as it ascends in the posterior mediastinum and arches forward to join the superior vena cava (SVC). The right paratracheal region contains a normal size node. The internal mammary arteries and veins are clearly seen in the parasternal regions bilaterally. C, Just below the carina, the right pulmonary artery is crossing the mediastinum. The azygoesophageal recess can be identified, with the lung contacting pleural overlying the esophagus and azygos vein. D, Right superior and left inferior pulmonary veins can be seen entering the left atrium (LA). E, At this level, the right ventricle (RV), right atrium (RA), left ventricle (LV) and left atrium (LA) can be clearly identified.

    Figure 1-3

    Computed tomography lung windows demonstrating normal anatomy. A, At this level, the right upper lobe bronchus can be seen along its length as it originates from the right main stem bronchus. The major fissures separate the upper lobes anteriorly from the lower lobes posteriorly. B, The bronchus intermedius is visible as an oval lucency. Its posterior wall should be smooth and thin. The left main and upper lobe bronchus can also be identified. The minor fissure is seen as a relatively avascular region. C, Image through the basal segmental bronchi of the lower lobes showing both the bronchi and their associated vessels.

  • The majority of CT scanners are now multislice machines with multiple rows of detectors (currently ranging from 4 to 256) within the doughnut-shaped gantry. The patient lies on the scanner table and is moved through the gantry as it continuously rotates, with each of the rows recording information. A large volume of data can now be acquired extremely quickly. There is a degree of tradeoff between resolution of the images and the speed of scanning. Most patients can hold their breath for 20 seconds, the approximate time required to image the thorax. If very thin slices are required for improved resolution, the scan will take longer to perform.

  • Scanners reconstruct sets of images in the axial plane for routine clinical work. Volume data acquisition allows for high-quality two- and three-dimensional reformations along multiple planes without additional radiation exposure. These additional reconstructions are helpful for diagnosis and surgical planning. Surface rendering is a particular form of three-dimensional imaging that is particularly useful for visualizing anatomic structural lumens such as those of the bronchial tree ( Fig. 1-4 ).

    Figure 1-4

    Images demonstrate axial ( A ), two-dimensional multi-planar reconstruction (MPR) ( B ), three-dimensional ( C ), and endoluminal ( D ) reconstructions of a standard thoracic computed tomography volume data set.

  • Tissues are displayed in terms of their attenuation of the x-ray beam. A small volume of tissue, a voxel, is displayed on a CT image as a two-dimensional pixel. The pixel is given a Hounsfield unit (HU) value depending on the density of its contained tissues. Water has a HU of 0, dense bone up to +2000 and air -1000. The Hounsfield unit values of the multiple pixels in the image are then mapped to a gray scale. Windowing is the process used to display the data to the full advantage. For image display purposes, no single window setting can adequately show all of the information available on a chest CT scan. The display of the CT image on the monitor is determined by the window width and level. Typically for thoracic CT examinations, one setting is used to optimally display the lungs and a second setting is used for the soft tissues.

Iodinated Contrast

  • Thoracic CT scans are often performed to specific protocols and can be performed with and without iodinated intravenous and oral contrast. Intravenous contrast is required in patients with suspected vascular abnormalities and is typically used for staging of carcinomas and in patients with suspected mediastinal, hilar, or pleural abnormalities.

  • Intravenous contrast administration is not without risk, and although adverse reactions occur at low rates, reactions are not infrequently encountered given the widespread use of iodinated contrast. Reactions range from mild urticaria to anaphylactic shock and death. The incidence of severe reactions with nonionic agents is 0.4% and of very severe reactions is 0.004%. Patients with an increased risk of contrast reaction include those who have had a previous contrast reaction, those with asthma, and those with multiple, well-documented allergies. In these individuals, unenhanced CT or a different imaging modality may suffice diagnostically.

  • Some institutions advocate premedication with steroids and antihistamines before further contrast administration, but there is no conclusive evidence of benefit for the prophylactic use of steroids in the prevention of severe reactions to contrast media.

  • Iodinated contrast media has nephrotoxic potential but rarely cause significant renal failure in patients with normally functioning kidneys. Contrast should be avoided in patients with existing renal impairment, with diabetes, with congestive heart failure of class III or IV, in patients with reduced effective arterial volume (e.g., nephrotic, cirrhotic), or those receiving drugs that may impair renal function or increase contrast nephrotoxicity.

Radiation Dose

  • Radiation exposure should be kept as low as reasonably achievable (ALARA). CT has made an enormous impact on diagnostic imaging; however, the radiation dose for CT is far greater than for conventional radiographs. In addition, the use of CT has markedly increased over time.

  • In a recent United Kingdom survey, CT scans constituted 7% of all radiologic examinations, but contributed 47% of the total collective dose from medical x-ray examinations in 2000/2001; in the United States, CT accounts for approximately 11% of all radiology procedures and almost 70% of the total effective radiation dose. Increased CT use has led to an overall increase in the total amount of medical radiation used, despite reductions in other areas.

  • The increase in radiation exposure via CT scanning may represent a public health issue in the future. A recent report in the New England Journal of Medicine suggested that the radiation from current CT scan use may cause as many as 1 in 50 future cases of cancer. The risks are greater in children because of their increased radiosensitivity and because of the greater years of life during which a radiation-induced cancer could develop.

  • Possible strategies to address the above-mentioned issues include decreasing the absolute number of CT examinations by using other imaging modalities such as ultrasound and magnetic resonance imaging (MRI) and altering scan parameters to decrease the dose of individual examinations. Education of patients and physicians regarding radiation dose and potential risks is critical in order to address the issue of increasing radiation exposure.

Positron Emission Tomography

  • Positron emission tomography (PET) and CT scanners can be combined in the same gantry to optimize the localization of PET abnormalities ( Fig. 1-5 ).

    Figure 1-5

    Positron emission tomography (PET)/computed tomography (CT) images in a patient with a right upper lobe non–small cell lung carcinoma with right hilar and mediastinal nodal metastases. CT ( bottom left ) and PET ( top right ) images are fused ( top left ) to provide an image with both anatomical and physiological information. Note the normal cardiac uptake and FDG activity in the bladder.

  • Patient preparation is required for PET. High glucose levels can compete with fluorodeoxyglucose (FDG) uptake and can degrade image quality; patients are typically not permitted to eat or drink for approximately 4 hours before the scan. Diabetics require special preparation. Caffeine, nicotine, and alcohol should be avoided for 24 hours before the scan. Oral hydration may be helpful.

  • Appropriate timing of the PET examination following procedures is required to prevent false-positive studies. The following guidelines are suggested :

    • Post biopsy 1 week

    • Post surgery 6 weeks

    • Post chemotherapy 4 to 6 weeks

    • Post radiation 4 to 6 months

  • Solitary pulmonary nodules

    • 1.

      PET scanning is well established for the evaluation of solitary pulmonary nodules with an intermediate pretest probability of malignancy.

    • 2.

      If the pretest probability is high, biopsy or resection should be considered. If the pretest probability is low, observation should suffice.

    • 3.

      A negative PET scan indicates that a nodule is highly likely to be benign, although it is important to take the pretest probability of malignancy into account.

    • 4.

      A positive PET scan indicates that the lesion is most likely to be malignant.

      Although false-positive results can occur, PET has higher specificity than CT or MRI scanning alone.

    • 5.

      A negative PET scan is usually more accurate than a positive PET scan.

      If the lesion does prove to be malignant, PET is the most accurate modality for staging mediastinal lesions (see Fig. 1-5 ).

    • 6.

      False-positive results have been described in active granulomatous/inflammatory disease. False-negative results occasionally occur in patients with bronchioloalveolar carcinoma, carcinoid, and in small (less than 1 cm) nodules. Therefore, PET scanning is usually reserved for lesions larger than 1 cm in size. PET can be considered in lesions smaller than 1 cm, but negative nodules should be closely observed.

Magnetic Resonance Imaging

  • MRI scanners generate a strong magnetic field around the patient, who is then exposed to short bursts of radio waves called radiofrequency (rf) pulses. The motion of protons in the body is altered by the rf pulses, which when they return to their original states, emit rf signals of their own, which are detected by the scanner and are used to construct a cross-sectional image of a slice of tissue.

  • The main advantage of MRI over CT is the lack of ionizing radiation and the superior soft tissue contrast resolution. The ability of MRI to produce images in multiple planes is less of a differentiating factor because excellent multiplanar images can now be reconstructed with multidetector computed tomography (MDCT).

  • MRI is not generally useful for imaging the pulmonary parenchyma but can provide excellent images of the mediastinum and chest wall. It is particularly useful for imaging superior sulcus tumors.

  • Nephrogenic systemic fibrosis (NSF), a rare multisystemic fibrosing disorder, has recently been described in patients receiving gadolinium-based MRI contrast agents. Initially termed nephrogenic fibrosing dermopathy because of the skin involvement, the disease has been shown to demonstrate systemic fibrosis involving skeletal muscle, bone, lungs, pleura, pericardium, myocardium, kidneys, testes and dura. Although the etiology of NSF is unclear, several contributing factors have been described, including renal insufficiency, major tissue injury such as arterial or venous thrombosis or surgery, and exposure to gadolinium-based contrast agents. At present, dialysis after the administration of gadolinium-based agents or a lower dose of agent has not been shown to decrease the incidence of NSF. The US Food and Drug Administration issued a public health advisory regarding gadolinium-based contrast agents in patients with renal failure in June 2006, which was updated in May 2007. The use of gadolinium-based contrast agents should be avoided unless the diagnostic information is not available with non–contrast-enhanced MRI in patients with acute or chronic severe renal insufficiency (glomerular filtration rate <30 mL/min/1.73m 2 , or acute renal insufficiency of any severity due to hepato-renal syndrome or in the perioperative liver transplantation period.

  • Contraindications to MRI scanning are:

    • Metallic foreign bodies in the eye.

    • Implanted electronic devices such as pacemakers.


  • Ultrasound uses high-frequency sound waves to image tissue. Reflected echoes are detected and used to create an image.

  • Thoracic ultrasound is most useful for the detection and characterization of pleural disease, particularly pleural effusions ( Fig. 1-6 ), and is often helpful in guiding thoracentesis and drain placements.

    Figure 1-6

    A, Free-flowing effusion and fluid with septa. Thoracic ultrasound of a free-flowing pleural effusion. The fluid (F) is hypoechoic ( black ) adjacent to the diaphragm ( black arrows ). A small triangular tongue of collapsed lung is also visible ( white arrow ). B. Thoracic ultrasound of a complex effusion containing fine septae ( arrows ).

  • In selected circumstances, peripheral lung lesions may be localized and biopsied using ultrasound for guidance.

  • Ultrasound may be useful in evaluating the diaphragm in cases of suspected diaphragmatic paralysis.

  • Ultrasound may also be used to assist in placement of central venous lines.

Esophageal Imaging

  • The esophagus can be imaged using barium esophagrams, CT, MRI, and endoscopic ultrasound. Barium esophograms provide useful information about mucosal surface lesions and esophageal motility. CT does not show the mucosa in detail but demonstrates tissues surrounding the esophagus and can evaluate extension of tumor and the presence of lymph nodes. Endoscopic ultrasound is the investigation of choice for the T staging of esophageal tumors but is still not universally available.

  • Esophageal trauma is usually diagnosed on the basis of chest radiographs and contrast swallows. If clinical evaluation and traditional radiographic studies are equivocal, CT should be performed.

  • Esophageal endoscopic ultrasound (EUS) may also be useful for that staging of lung cancer. In particular, lower level 7 (subcarinal), 8 (paraesophageal), and 9 (inferior pulmonary ligament) lymph nodes may be better visualized or biopsied by EUS than by endobronchial ultrasound (EBUS) or mediastinoscopy.


General Principles

Knowledge of the lymphatic drainage pattern in the thorax is useful in the staging of a variety of thoracic neoplasms, and for the accurate assessment of nodal involvement which is a critical factor in staging of malignancy.

Nodal short axis has been shown on autopsy studies to be a more accurate predictor of nodal size than long axis. Although the size of normal nodes may vary depending on the region of the medistinum, in general, lymph nodes in the paratracheal, aortopulmonary window, hilar, subcarinal, and paraesophageal regions are considered abnormal if the nodal short axis is greater than 1 cm. Peridiaphragmatic and internal mammary nodes are considered abnormal if the nodal short axis is greater than 5 mm. Nodes within the retrocrural and extrapleural regions are not normally visible, and should be considered abnormal when present. However, accurate staging requires pathologic examination of enlarged nodes, because nodal size alone is not reliable in the assessment of neoplastic involvement.

Imaging Principles in Lung Cancer

  • Accurate staging is critical in determining the most appropriate therapy and prognosis. Familiarity with the current staging system for lung cancer is a requirement for those involved in the assessment of these patients.

  • The goal of the surgeon is to surgically resect all patients with a resectable tumor who are medically able to tolerate resection, yet avoid unnecessary thoracotomies.

  • The radiologist is critical in helping to determine whether the patient is resectable with accurate descriptions of T status to include primary tumor size and extent; N status to include location, size, and number of nodes; and M status to include presence and location of metastases.

  • Roadmapping of the primary tumor, nodes and metastases can be performed to aid biopsy ( Fig. 1-7 ).

    Figure 1-7

    A, The frontal radiograph demonstrates a large, irregularly marginated, mass in the right upper lobe with some adjacent distortion of the superior right hilum. B to C, CT images confirm the presence of a thick-walled, irregularly marginated, cavitating tumor in the right upper lobe. D, Right hilar and low right paratracheal nodes are greater than 1 cm in short axis diameter ( arrows ). The right paratracheal nodes were positive for malignancy at mediastinoscopy precluding surgery.

  • Radiographic features that indicate unresectability or stage IIIB or IV disease include T4 disease with invasion of mediastinum or diaphragm, ipsilateral pleural metastatic disease, N3 disease with contralateral mediastinal, contralateral or ipsilateral supraclavicular nodes, and M1 disease with distant metastases.

Primary Tumor

  • The main role of CT in the assessment of the primary tumor is in differentiation of T3 from T4 lesions.

  • A CT diagnosis of T4 tumor can be made if

    • 1.

      Tumor is involving the trachea or narrowing the carina.

    • 2.

      Tumor is surrounding, distorting, or attenuating, or having greater than 180 degrees of contact with the superior vena cava, the aorta, the main pulmonary artery, right or left pulmonary artery within the mediastinal pleural reflection, or the central pulmonary veins.

    • 3.

      Tumors abutting the superior vena cava with elevation of the diaphragm to indicate invasion of the phrenic nerve.

    • 4.

      Tumor results in the destruction of a vertebral body ( Fig. 1-8 ) or involvement of the brachial plexus.

      Figure 1-8

      Computed tomography image of a large T4 tumor in the posterior left upper lobe invading the adjacent vertebral body.

    • 5.

      The presence of pleural carcinomatosis as confirmed on CT by soft tissue pleural nodules.

  • In selected cases, CT or MRI can reliably identify involvement of major thoracic structures, that is, T4 disease, precluding surgery ( Figs. 1-9 and 1-10 ). Findings suggesting mediastinal invasion include extensive contact between the tumor and the mediastinum, loss of a fat plane between tumor and the mediastinum, mass effect on adjacent mediastinal structures, and pleural and pericardial thickening.

    Figure 1-9

    Computed tomography image of an unresectable T4 tumor that is invading the left atrium via the left inferior pulmonary vein ( arrows ).

    Figure 1-10

    Magnetic resonance imaging scan demonstrates filling defect within the left atrium (LA) from direct tumor (T) invasion.

  • The assessment of mediastinal invasion can however be very difficult using CT alone because inflammation and desmoplastic reaction can simulate tumor invasion, and microscopic extension may be missed. Minimal invasion of pericardium, fat, vagus, and phrenic nerves are generally considered resectable disease. Radiologic determination of the proximity of lesions to the tracheal carina may also be problematic, although the ability to view CT reconstructions in multiple planes is often very helpful.

  • Chest wall invasion does not rule out surgery unless there is invasion of the subclavian artery or of a vertebral body. CT is limited in assessing chest wall invasion, with reported sensitivities and specificities ranging from 38-87% and 40-89%, respectively. The only reliable findings of chest wall invasion are the presence of rib destruction and a chest wall mass ( Fig. 1-11 ). However, these findings are present in only 20% to 40% of patients with surgically proven chest wall invasion. Although the sensitivity and specificity of MRI is greater than those of CT, MRI is also of limited value. When a lung tumor abuts the parietal pleura, the presence of chest wall pain is a better predictor of chest wall invasion than the appearance on the CT scan, unless there is rib destruction.

    Figure 1-11

    Computed tomography image of a large left sided tumor that is invading the chest wall, producing an obvious chest wall mass ( asterisk ).

  • Invasive tumors arising in the lung apex may produce the characteristic clinical findings of Horner’s syndrome and pain involving the shoulder and arm. CT can provide anatomic information regarding the local extent of the tumor, and demonstration of this region has improved with the multiplanar views made possible with multislice CT. The superior contrast resolution of MRI makes it the preferred modality for assessment of the extent of local tumor invasion, particularly with respect to brachial plexus and vascular involvement ( Fig. 1-12 ).

    Figure 1-12

    Coronal ( A ) and sagittal ( B ) MRI scans of the thoracic inlet show a rounded mass at the right lung apex that is invading the chest wall at the right lung apex ( asterisk ).

Nodal Disease

  • Prognosis in lung cancer is influenced by the number and size of nodes, and intracapsular versus extracapsular and microscopic versus macroscopic disease.

  • CT is limited in its ability to stage mediastinal nodes accurately because it relies on nodal size alone. Nodes are considered enlarged when they measure greater than 10 mm in short axis. However, benign hyperplastic nodes may be enlarged, and normal size nodes may contain microscopic foci of malignant disease. In a study by McLoud and colleagues, 13% of nodes less than 1 cm contained metastases, and nearly one third of nodes 2 to 4 cm were hyperplastic nodes and did not contain metastases. As such, the most important role of CT in the assessment of nodal disease is in determining the presence of enlarged nodes, and CT is the best method of directing nodal sampling for histological evaluation ( Fig. 1-13 ). The sensitivity and specificity for the CT evaluation of mediastinal nodal metastases are both about 65%.

    Figure 1-13

    Contrast-enhanced computed tomography image demonstrating enlarged subaortic (station 6 nodes). The short axis diameter of these nodes (lines) measured greater than 1 cm.

Metastatic Disease

  • Extrathoracic metastases are present in approximately 40% of patients with newly diagnosed lung cancer ( Fig. 1-14 ).

    Figure 1-14

    A and B, M1, stage IV lung carcinoma with metastases. A, Computed tomography (CT) lung window demonstrating a large right upper lobe tumor with multiple round nodules bilaterally, which is consistent with metastases. B, CT soft tissue image through the upper abdomen showing a large, heterogeneous attenuation, adrenal metastasis ( asterisk ).

  • Clinical evaluation and laboratory tests are usually performed to evaluate for metastatic disease; the use of CT in the assessment of extrathoracic metastases in asymptomatic patients with normal biochemistry remains controversial.

Positron Emission Tomography in the Staging of Non–Small Cell Lung Carcinoma

  • PET is now standard for staging of non–small cell carcinoma of the lung. The addition of PET to the conventional workup of lung cancer patients will prevent unnecessary surgery in one out of five patients and will change the stage of more than half of the patients. PET mediastinal staging has a reported sensitivity of 81%, and specificity 90%; PET/CT has superior accuracy compared with that of CT or PET alone ( Fig. 1-15 ).

    Figure 1-15

    Computed tomography (CT) and fused positron emission tomography/computed tomography (PET/CT) images of a non–small cell lung cancer in a 45-year-old woman. A, CT lung window demontrates the primary lesion in the right upper lobe. CT soft tissue image ( B ) shows enlarged nodes in the low right paratracheal region ( arrow ) and ( C ) small (less than 1 cm) short axis diameter nodes in the high right paratracheal region and in the contralateral mediastinum ( arrows ). Fused PET/CT ( D & E ) images reveals avid fluorodeoxyglucose uptake in both the enlarged right paratracheal nodes but also in the smaller nodes, upstaging the patient from N2 to N3.

  • PET is useful in patients with potentially operable disease and no abnormal mediastinal nodes or distant metastases on CT. If the PET scan is negative, mediastinoscopy can potentially be avoided. In patients with enlarged mediastinal nodes, the false-negative rate of PET is approximately 5% to 9%, compared with an estimated false-negative rate for mediastinoscopy of 9%. Understanding the standard pattern of lymphatic spread of lung lesions by location is helpful in avoiding false-positive results both in CT and PET scanning.

  • Positive PET results should be confirmed by mediastinoscopy or lymph node sampling before excluding surgery, because the false-positive rate of PET in the mediastinum is 13% to 22%. Mediastinoscopy may be avoided if the pretest probability is very high. However, patients who have dual pathology such as tuberculosis may have false-positive results.

  • PET detects unsuspected distant metastases in approximately 10% of patients. PET has equal or greater sensitivity and is more specific for the diagnosis of bone metastases from lung carcinoma, and 100% sensitivity and 80% specificity for adrenal metastases has been reported. Adrenal hyperplasia may have increased uptake and can mimic bilateral adrenal metastases in patients with carcinoid or small cell lung carcinoma. However, PET has only a 60% sensitivity for brain metastases due to the high background metabolic activity of the brain and, therefore, cannot be used in place of CT or MRI for evaluation of brain metastases.

  • PET can be used to differentiate malignant from benign pleural disease with a sensitivity of 97% and a specificity of 89%. Intense uptake is highly predictive of malignancy; moderate uptake should be interpreted with caution because inflammatory disorders may result in this appearance. A prior talc pleurodesis can also cause increased pleural uptake.

  • PET can be helpful in the post-thoracotomy chest for diagnosing recurrent malignant disease. PET can differentiate between local recurrence and post-treatment change with a sensitivity of 97% to 100% and a specificity of 62% to 100%. False-positive results may occur from inflammatory reactions following therapy. PET is more accurate than CT for the detection of local disease recurrence, and PET/CT is substantially more specific than PET alone.

Imaging Principles in Mesothelioma

  • Mesothelioma is the most common primary neoplasm of the pleura but is still a relatively rare disease. The vast majority of cases are secondary to asbestos exposure. Presenting symptoms include chest pain and shortness of breath.

  • The presence of unilateral effusion, volume loss, nodular pleural thickening, and pleural thickening involving mediastinal or fissural pleura suggests mesothelioma ( Fig. 1-16 ). Calcified plaques suggesting asbestos exposure are seen only in the minority of mesothelioma patients. CT is currently the imaging investigation of choice for diagnosis and staging of this disease. MRI and PET scanning have useful complimentary roles in making treatment decisions.

    Figure 1-16

    A, Posteroanterior chest radiograph in a patient with mesothelioma. There is circumferential, nodular, pleural thickening and volume loss of the left hemithorax. B, Contrast-enhanced axial computed tomographic images confirm the presence of circumferential thickening of the pleura. The mediastinal pleura is involved, and the pleural thickening exceeds 1 cm; both are indicators of pleural malignancy. C, Similar findings of pleural mass/thickening at lower level.

  • Mesothelioma is staged using a system from the International Mesothelioma Interest Group. This system emphasizes criteria used to determine the extent of local tumor and lymph node involvement.

  • It is important to distinguish between the potentially resectable T3 tumor and the unresectable T4 tumor. As in non–small cell lung carcinoma, the presence of N3 nodal disease and distant metastases precludes surgery.

  • Mesothelioma is locally aggressive and can involve intercostal muscles and adjacent bone. It can extend into the chest wall along chest tube tracts and biopsy sites.

  • Pericardial involvement may be seen at CT as nodular pericardial thickening or pericardial effusion. Chest CT may demonstrate pulmonary metastases or rarely extrathoracic spread.

  • Metastases to lymph nodes are present at autopsy in approximately 40% to 45% of patients. As in lung carcinoma staging, the evaluation of lymph nodes by CT has a relatively low sensitivity and specificity because it relies on size alone. CT can also underestimate chest wall involvement and peritoneal disease.

  • The superior contrast resolution of MRI may allow for improved detection of tumor extension, especially to the chest wall and diaphragm.

  • PET can distinguish mesothelioma from benign pleural processes and can be useful in staging and preoperative evaluation of the disease. Areas of high standardized uptake value (SUV) within diffuse pleural disease may guide biopsy. PET has increased accuracy in the detection of mediastinal nodal metastases when compared with CT. PET may predict prognosis in that higher FDG activity is associated with a poorer prognosis. A high level of uptake is associated with the presence of N2 disease and poor outcome. PET is very useful in the identification of distant metastases that preclude thoracotomy.

Imaging Principles in Esophageal Carcinoma

  • Cross-sectional imaging is used to stage esophageal cancer but is somewhat suboptimal. Current T staging emphasizes the depth of invasion of the primary tumor. CT and MRI scanning cannot reliably delineate the individual layers of the esophageal wall and therefore cannot be used to differentiate between T1 and T2 lesions. Accurate T1/T2 staging relies on esophageal ultrasound, which has an overall accuracy of 85% to 95%.

  • The accuracy of CT in identifying mediastinal invasion varies between 59% and 82%. Loss of intervening fat planes may be helpful. If the esophageal mass is displacing or indenting the posterior wall of the trachea or bronchus, the tumor is probably invading the structure.

  • Involvement of the aorta or tracheobronchial tree signifies inoperable, T4 tumor.

  • CT findings include

    • 1.

      Irregular thickening of the esophageal wall

    • 2.

      An intraluminal polypoid mass

    • 3.

      Eccentric narrowing of the lumen

    • 4.

      Dilatation of the esophagus proximal to the narrowing

    • 5.

      Tumor invasion of adjacent structures

  • Nodal spread is typically to paraesophageal, other mediastinal, gastrohepatic ligament, and left gastric nodes ( Fig. 1-17 ). As in the staging of other carcinomas, CT is limited by its reliance on anatomic size of the nodes. It can be difficult to differentiate enlarged paraesophageal nodes from contiguous tumor spread.

Jun 24, 2019 | Posted by in CARDIAC SURGERY | Comments Off on Imaging of the Thoracic Surgery Patient

Full access? Get Clinical Tree

Get Clinical Tree app for offline access