Ten to 30 percent of all fiberoptic bronchoscopy examinations are performed in patients with hemoptysis, primarily to exclude cancer as a cause. A thorough evaluation is clearly indicated for a patient whose chest roentgenogram shows localizing findings suspicious of lung cancer. However, there has been debate concerning whether the risk of cancer is high enough to warrant bronchoscopy in patients with hemoptysis and a nonsuspicious chest roentgenogram, ie, one which is either normal or shows only nonlocalizing abnormalities. After reviewing the literature on this topic, Jackson et al concluded that the risk of cancer was negligible in patients less than 40 years old but that additional data were needed to determine which patients more than 40 years old required bronchoscopy. more
Some authors have found a high frequency of cancer (more than 15 percent) in older patients with hemoptysis and a nonsuspicious chest roentgenogram, while others have found no cancers and have concluded that bronchoscopy is unnecessary. Possible explanations for this discrepancy include small sample sizes, differences in patient characteristics or referral patterns, and inclusion of patients with nonlocalizing abnormalities in addition to those with normal chest roentgenograms.
When evaluating the role of bronchoscopy in this setting, it is important to consider not only the frequency with which cancers are detected, but also the likelihood that the cancers detected will be treatable. Surgical excision is generally considered to be the only effective treatment for non-small cell lung cancers. However, the frequency with which the presumably central tumors causing hemoptysis and a nonsuspicious chest roentgenogram are resectable is not known.
In this study, we reviewed the Minneapolis Veterans Administration Medical Center experience with fiberoptic bronchoscopy in older men with hemoptysis and a nonsuspicious chest roentgenogram. Our purpose was to determine the rate of cancer detection as well as the yield of treatable disease (resectable cancers or other treatable conditions) identified by bronchoscopy. We also sought to determine whether nonlocalizing abnormalities on chest roentgenogram, though not in themselves suggestive of cancer, nevertheless conferred an increased risk compared with a chest roentgenogram read as “normal,” as has been previously suggested. Finally, we wished to evaluate the conclusion of Adelman et al that a negative bronchoscopic examination effectively excludes cancer in patients with hemoptysis and a nonsuspicious chest roentgenogram.