Basic methods of diagnosis of diseases of the chest are skin tests, endoscopy, mediastinotomy, video-assisted thoracoscopic surgery, biopsy, CT, MRI, PET, sputum analysis.
Skin tests are used in the diagnosis of tuberculosis, histoplasmosis, and coccidioidomycosis. Tuberculin testing is usually done with purified protein derivative injected intradermally. Intermediate-strength PPD should be used in patients who seem likely to have active disease. Induration of 10 mm or more at the injection site after 48–72 hours is called positive and indicates either active or arrested disease. Because false-negative reactions are rare, a negative test fairly reliably rules out tuberculosis. Mumps antigen is usually placed on the opposite forearm to test for anergy. Skin tests for histoplasmosis and coccidioidomycosis are performed in a similar way, but skin tests for fungal infections are unreliable and serologic tests should be performed instead.
Indirect laryngoscopy is used to assess vocal cord mobility in patients suspected of having lung carcinoma, especially when there has been a voice change. It should be performed also to search for an otherwise occult source for malignant cells in sputum or metastases in cervical lymph nodes.
Roentgenographic evidence of bronchial obstruction, unresolved pneumonia, foreign body, suspected carcinoma, hemoptysis, aspiration pneumonia, and lung abscess are only a few of the indications for bronchoscopy. The procedure can be done using either the standard hollow metal (rigid) or the flexible fiber optic bronchoscope under local anesthesia. Rigid bronchoscopy must be done under general anesthesia, is most often used for clearing major airways of bulky obstructing lesions such as tumors, foreign bodies, or blood clots. Traditionally the CO2 laser required use of the rigid bronchoscope, though more recently developed technology can be applied with flexible bronchoscopy.
Flexible bronchoscopy is a highly effective diagnostic and therapeutic tool. It can be performed under local and intravenous sedation. Washings are usually obtained for bacterial or fungal culture and cytologic examination. Visible lesions are biopsied directly, biopsy specimens are sometimes taken from the carina even though it appears normal. Brush biopsies are obtained from specific bronchopulmonary segments. Occasionally, transcarinal needle biopsy of a subcarinal node is obtained. Thirty to 50 percent of lung tumors are visible bronchoscopically. Brushing, random biopsies, and sputum cytology may still yield a positive diagnosis of cancer or tuberculosis in the absence of a visible lesion. The yield is influenced by size, location, and histologic cell type of the lesion.
Cervical mediastinoscopy remains a mainstay of evaluation of the mediastinum despite advances in imaging. Properly performed mediastinoscopy samples nodes from at least three stations, including ipsilateral and contralateral paratracheal levels. Cervical mediastinoscopy is performed through a 3- to 4-cm incision one fingerbreadth above the sternal notch. Dissection proceeds beneath the pretracheal fascia, allowing safe access to mediastinal nodes and avoiding major vascular structures. After palpation, the mediastinoscope can be inserted and nodes biopsied under direct vision. Unclear structures can be aspirated prior to attempted biopsy.
Enlarged lymph nodes in the aorticopulmonary window are technically inaccessible by means of standard cervical mediastinoscopy. Extended cervical mediastinoscopy, however, is a technique that provides access to these aorticopulmonary window nodes. It is performed through the same small neck incision as standard mediastinoscopy except the dissection is carried laterally beside the left carotid artery toward and then over the aorta into the aorticopulmonary space. In older patients with densely calcific aortas, extended mediastinoscopy is contraindicated because of the risk of embolic phenomena and stroke from aortic manipulation.
In experienced hands, the complications of mediastinoscopy are minimal (< 1–2%). Major bleeding complications requiring sternotomy or thoracotomy for repair are infrequent (1–2%). Other possible complications include pneumothorax, recurrent nerve injury, infection, and esophageal injury.
Mediastinoscopy is almost invariably accurate in the diagnosis of sarcoidosis. It is also useful to diagnose tuberculosis, histoplasmosis, Castleman's silicosis, metastatic carcinoma, lymphoma, carcinoma of the esophagus. It should not be used in the investigation of primary mediastinal tumors, which should be approached by an incision permitting definitive excision.
Anterior mediastinotomy (the Chamberlain procedure) is used to sample nodes and biopsy tissue in the anterior mediastinum, most commonly in the aorticopulmonary window. A small (3- to 4-cm) incision is made over the second or third interspace on the appropriate side of the lesion. Alternatively, the procedure can be performed with videoscopic guidance (VATS). The mediastinum is approached through the interspace directly or after excising the costochondral cartilage using either the mediastinoscope or open technique. Careful attention is paid to preserving the mammary vessels encountered in the dissection. The mediastinum is approached extrapleurally unless lesions specifically within the thorax—effusions, tumors invading the hilum or chest wall—need to be investigated. Furthermore, if additional access is required to facilitate the dissection or to treat a complication, the incision can be converted easily to a larger anterior thoracotomy.
Complications resulting from anterior mediastinotomy are similar to those encountered with cervical mediastinoscopy and include bleeding, recurrent nerve injury, and infection. Major morbidity should be less than 2%.
Video-Assisted Thoracoscopic Surgery
With the advent of sophisticated video technology and optics, video-assisted thoracoscopic surgery has evolved as an important tool in thoracic surgery. While thoracoscopic procedures have been used for many years, newer technical advances as well as increasing surgeon familiarity with minimally invasive surgery have made VATS increasingly popular and useful.
VATS plays an important role in the diagnosis, staging of thoracic malignancies (lung cancer, mesothelioma, etc) as well as in the resection of isolated peripheral pulmonary nodules and bullous lung disease. Furthermore, it has been an advance in lung biopsy and pleurodesis procedures.
However, despite gaining popularity, many thoracic surgeons consider the approach suboptimal for lung cancers. Full mediastinal lymph node dissections are not generally obtainable with standard VATS. Even for metastasectomy, concerns over pleural and chest wall seeding with VATS have been documented, highlighting the need for assiduous attention to detail during these procedures. The improved resolution from new-generation CT scanners have in part supplanted the need for palpation of the entire lung, and increasingly advanced resections are now being performed via VATS approaches. At present VATS is most commonly applied to benign processes, including spontaneous pneumothoraces and pleural effusions, as well as for thoracic sympathectomy and thoracic vertebral discectomy.
As instruments and techniques have evolved, complications from VATS procedures — persistent air leaks, hemorrhage, tumor seeding, etc — have decreased. Overall, major complication rates of 1–2% are reported. Faster patient recovery, shorter hospital stays, decreased pain are major advantages of VATS, although long-term differences between videoscopic and formal thoracotomy using muscle-sparing incisions are yet to be demonstrated.
Scalene Lymph Node Biopsy
Biopsy has been largely replaced by mediastinoscopy in the evaluation of pulmonary disease since it offers the same information but is less reliable and does not evaluate nodes within the mediastinum. Furthermore, scalene lymph nodes are accessible via cervical mediastinoscopy using the mediastinoscope. In the evaluation of lung cancer, about 15% of scalene node biopsies are positive when the cervical nodes are not palpable compared with 85% when the nodes are palpable. The risk of major complications is about 5%. Deaths are rare.
Needle biopsy is indicated when the cause of a pleural effusion cannot be determined by analysis of the fluid or when tuberculosis is suspected. Any one of three needles can be used: Vim-Silverman, Cope, or Abrams (Harefield). A definitive diagnosis can be obtained in 60% of cases of tuberculosis or cancer. The principal complication is pneumothorax. Five to 10 percent of biopsy specimens are inadequate for diagnosis.
Biopsy of the pleura can be performed via videoscopic or open technique with minimal morbidity, providing the pathologist with a specimen superior to that of needle biopsy.
The indications for percutaneous needle biopsy are not well established. It may be indicated in diffuse parenchymal disease, in some patients with localized lesions. The diagnosis of interstitial pneumonia, carcinoma, sarcoidosis, hypersensitivity lung disease, lymphoma, pulmonary alveolar proteinosis, and miliary tuberculosis has been established by this method.
Needle biopsies are done by any of three techniques: by aspiration with a cutting needle, by trephine, or by air drill. Needle biopsy of the lung is also possible by a transbronchial technique in which a modified Vim-Silverman or ultrathin needle is used.
There is controversy concerning the risks of spreading the tumor by needle biopsy in localized disease. Complications following percutaneous needle biopsy include pneumothorax (5–30%), hemothorax, hemoptysis, and air embolism. Pulmonary hypertension or cysts and bullae are contraindications. Several deaths have been reported. There is about a 60% chance of obtaining useful information.
Thoracoscopy or VATS has become the procedure of choice for open biopsy in patients who can tolerate single lung ventilation. A single anterior axillary line incision can be used for introduction of a stapler and operating thoracoscope. For open biopsies a limited intercostal or anterior parasternal incision is used to remove a 3- to 4-cm wedge of lung tissue in diffuse parenchymal lung disease. The site of incision is selected for accessibility and potential diagnostic value. The incision is generally made at the fifth interspace on the right at the anterior axillary line to allow for access to all three lobes for biopsy — or at the lower lobes bilaterally. The middle lobe, lingula are selected in specific cases when pathology exists only in these areas, as they generally yield results of the poorest quality. Open lung biopsy is associated with a lower death rate, fewer complications, and greater diagnostic yield than needle biopsy. It is especially useful in critically ill, immunosuppressed patients for differentiation of infectious infiltrative lesions from neoplastic infiltrative lesions. Open lung biopsy can be performed in the ICU setting and does not require single lung ventilation. When a focal lesion is biopsied, a larger incision is used. Peripheral lesions are totally excised by wedge or segmental resection, and deeply placed lesions are removed by lobectomy in suitable candidates.
Exfoliative sputum cytology is most valuable for detection of lung cancer. Specimens are obtained by deep coughing or by abrasion with a brush, or bronchial washings are obtained by either bronchoscopic or percutaneous transtracheal washing techniques. Specimens should be collected in the morning and delivered to the laboratory promptly. Centrifugation or filtration can be used to concentrate the cellular elements.
In primary lung cancer, sputum cytology is positive in 30–60% of cases. Repeated sputum examination improves the diagnostic return. Examination of the first bronchoscopic washing material yields a diagnosis in 60% of cases. Postbronchoscopy sputum analysis should always be made at 6–12- 24 hours, as findings may be positive at these times when previous tests were negative.
Newer technologies, including better sputum-inducing agents, are currently in clinical trials and appear effective. Also, more sophisticated cytologic analysis using immunohistochemistry to molecular markers (cytokeratins, hnRNP, etc) has improved accuracy to detect premalignant lesions.
Computed tomography is a cornerstone of evaluation of chest pathology. CT scanning is critical in the staging of carcinoma, and has value in defining the extent of metastatic disease.
Magnetic Resonance Imaging
Although the major value of MRI in the thorax has been in cardiovascular imaging, it has been moderately helpful in showing invasion of lung cancer into the chest wall, vertebrae, and spinal cord as well as mediastinal structures. MRI has a particular niche in the evaluation of superior sulcus tumors to establish involvement of the brachial plexus, subclavian vessel, or bony chest wall.
Position Emission Tomography
The presence of a positive PET scan in the mediastinum mandates either mediastinoscopy or, more recently, endoscopic evaluation of mediastinal lymph nodes because of false-positive PET scan results.
The combined PET/CT is highly accurate (> 90%), but by itself it has a 10–20% false-positive rate in the mediastinum. Therefore, interpretations of PET results must be accepted with caution and must be confirmed by staging when inconsistent with the overall clinical picture.