Lung
Chest x-ray films. Chest x-ray films have been the usual means of detecting lung cancer. Unfortunately, best results are obtained from the less common peripheral lesions rather than the more usual bronchogenic carcinomas arising in major bronchi. In general, chest x-ray films are not an efficient means of early diagnosis, and this is especially true for the miniature films used in mass survey work.
Sputum cytology. Sputum cytology is generally considered more sensitive than x-ray films, although some studies detected about equal numbers of asymptomatic tumor with either technique. Sputum cytology yield increases if the patient is a smoker and has symptoms such as chronic cough, hemoptysis, or recurrent pneumonia. Sputum samples for cytology should be obtained once daily (before breakfast and after rinsing the mouth with water) for at least 3 days. The material should be from a “deep cough”; saliva is not adequate. Many cytopathologists recommend expectoration directly into a bottle containing a special fixative (e.g., 50% ethanol, with or without additives). This type of specimen cannot be used for bacterial culture. In patients who do not have a productive (sputum-producing) cough, aerosol induction of sputum has been recommended. Some investigators achieved better results with a 3-day collection period than with aerosol inducement when several deep-cough specimens per day were expectorated directly into sputum cytology fixative. Twenty-four-hour collections without fixative are not recommended due to cell disintegration. A good specimen is the key to success in pulmonary cytology, because tumor cells may not be present continuously, because upper respiratory tract material usually does not reflect lower respiratory tract disease, and because pulmonary cytologic interpretation is more difficult than with uterine material.
Sensitivity of various diagnostic methods for lung cancer is not always easy to determine from the literature. The detection rate of carcinoma in asymptomatic persons (occult carcinoma) is naturally lower than in patients who have symptoms. For sputum cytology, better results are found if detection rates are used (“definitely positive” plus “suspicious” diagnoses) rather than only positive diagnoses. Unfortunately, it is often not clear which reporting method is being used. There is no question that more than one sputum sample, each sample being obtained on different days, significantly increases diagnostic yield (Table 33-12). Aerosol inducement of sputum also increases diagnostic yield (by about 20%–30%). There is a difference in detectability of central lesions (higher sputum cytology sensitivity) versus peripheral lung lesions (lower sputum sensitivity). For that reason, squamous cell carcinoma is more readily detectable by sputum cytology (literature range, 58%–85%), due to its tendency for proximal bronchus origin, than is adenocarcinoma (10%–57%), which tends to be peripherally located. Small cell undifferentiated carcinoma has intermediate detectability (30%–70%).
Table 33-12 Sensitivity of sputum cytology in primary lung carcinoma
Bronchoscopy. Bronchoscopy is reported to detect about 70%–80% of cases (45%–90%), with better detection of central versus peripheral lesions and with better results from direct biopsy of visible lesions versus bronchial washings, brushings, or blind biopsy. Percutaneous needle biopsy of lung tumors visible on x-ray film is reported to verify about 65%–70% of cases (48%–90%). Scalene node biopsy detects about 10% of cases (5%–21%). Mediastinoscopy with mediastinal node biopsy is reported to provide the only preoperative tissue evidence of carcinoma in 7%–20% of cases.
When cytologic material is obtained by bronchoscopy, saline is often used for bronchial washings. It is essential to use some type of “physiologic” saline rather than “normal” saline; contrary to common belief, the two are not identical. Normal saline (0.85% sodium chloride [NaCl]) can produce cellular artifact if the slides are not prepared within 5 minutes after collection. Physiologic saline is a balanced salt preparation with other minerals besides NaCl.
Radionuclide scans. When the diagnosis of lung cancer is first made, the question frequently arises as to which tests might help delineate extent of disease and thus establish operability. Bone scan is reported to detect lesions in approximately 35%–45% of patients, the frequency correlating roughly with clinical stage of the disease. However, a smaller number of those who are asymptomatic have an abnormal bone scan (14%–36%) and some of these abnormalities may not be due to metastasis. Brain scans have produced as many as 14%–20% positive results, but most studies found less than 6% were positive if there were no neurologic signs or symptoms. Initial workup liver scans disclose 13%–19% of abnormal results but less than 6% when there is no laboratory or physical examination evidence of liver metastases.
Computerized tomography. CT has been very useful to determine operability by visualizing the size and location of the lesion, the presence of thoracic metastases, and the size of the mediastinal lymph nodes.