Differential Diagnosis

A large number of disorders have been reported to cause hemoptysis, and the most important are listed in Box 20-1. Of these, bronchogenic carcinoma, bronchiectasis, bronchitis, and bacterial pneumonia are responsible for most cases. Table 20-1 shows the relative frequency of disorders causing hemoptysis in major series published since 1980. The significant variability, especially in the frequency of bronchiectasis, bronchitis, and tuberculosis, probably reflects differences in the time of publication, the patient population studied, and the diagnostic tests and criteria used. Figure 20-1 illustrates the percentage of patients with each diagnosis on the basis of pooled data from these studies.

Figure 20-1 Final diagnosis in patients presenting with hemoptysis. The approximate percentage of cases attributed to each diagnosis is shown.

Patient Evaluation

When the patient reports a history of expectorating blood, the first step must be to determine whether hemoptysis has actually occurred. That is, bleeding must be localized to the lower respiratory tract, and alternative sites, such as the nose, mouth, pharynx, larynx, and gastrointestinal tract, must be excluded. Few patients have difficulty distinguishing between vomiting and expectorating blood, although specific questions may be required to elicit a report of nausea and retching. Distinguishing between an upper and a lower airway source of bleeding occasionally is more difficult, although this usually can be accomplished by a directed history and physical examination. Patients with hemoptysis almost always report that the expectoration of blood follows an episode of coughing; in those with an upper airway source, it typically is preceded by a feeling of blood pooling in the mouth or the need to “clear the throat.” A history of epistaxis also is an important indicator of upper airway hemorrhage. Routine examination of the nose, mouth, and pharynx is important to rule out an obvious site of bleeding. A thorough examination that includes rhinoscopy and laryngoscopy is indicated when an upper airway source cannot be reliably excluded.

As shown in Table 20-3, the chest radiograph also may yield important information about the underlying cause of hemoptysis. The chest film, however, is “localizing”—that is, it demonstrates a mass, cavity, infiltrate, lobar atelectasis, or other finding that is likely to be directly related to the cause of hemoptysis—in less than 40% of patients. In the remainder, the chest radiograph is either normal in appearance or demonstrates abnormal but nonspecific findings such as emphysema, interstitial fibrosis, minor atelectasis, or pleural thickening—a category referred to as “nonlocalizing.” This radiographic classification has important diagnostic and prognostic implications. Malignancy is found in almost 40% of patients with hemoptysis associated with localizing findings on the chest radiograph. On the other hand, cancer is diagnosed in only 6% to 10% of patients with normal-appearing or nonlocalizing chest radiographs.

Table 20-3 Important Radiographic Findings in Patients with Hemoptysis

Additional Testing

The history, physical examination, and plain chest radiography are essential to reduce the number of possible causes of hemoptysis, and findings often point toward specific disorders. This initial evaluation, however, yields a definitive diagnosis in only a small percentage of patients. In most cases, additional testing is required, which most commonly consists of computed tomography (CT) and fiberoptic bronchoscopy (FOB).

Computed Tomography

Compared with conventional chest radiography, CT is clearly superior for imaging the peripheral and central airways, mediastinum, and lung parenchyma. Thus, it is not surprising that CT has been shown to be very useful in the evaluation of patients with hemoptysis. In the presence of a normal-appearing or nonlocalizing chest radiograph, CT reveals an unsuspected cause of hemoptysis, most commonly bronchiectasis, in approximately one third of patients (Figure 20-2). CT also may demonstrate an endobronchial lesion (Figure 20-3) or an unsuspected parenchymal mass, nodule, or cavity (Figure 20-4). CT also is useful in more than half of all patients whose chest radiograph is localizing, either by revealing a new source of hemoptysis or by providing additional information about a previously recognized abnormality.

Figure 20-2 Computed tomography (CT) appearance of bronchiectasis. Dilated peripheral airways are clearly seen on this high-resolution CT image.

Figure 20-3 Computed tomography (CT) scan showing an endobronchial mass. The lesion, which proved to be a bronchogenic carcinoma, is clearly visible in the right main bronchus.

Figure 20-4 Chest radiograph and computed tomography (CT) scan obtained in a patient with a cavitary squamous cell carcinoma. A, The lesion cannot be seen on the chest radiograph but is clearly demonstrated on the CT scan, shown in B.

Fiberoptic Bronchoscopy

Since becoming widely available in the early 1970s, fiberoptic bronchoscopy (FOB) has been used almost routinely in the evaluation of patients with hemoptysis. By combining endoscopic examination with brushings, washings, endobronchial and transbronchial biopsy techniques, and transtracheal needle aspiration, FOB may be used both to identify the site of bleeding and to obtain a definitive diagnosis. FOB is most useful for diagnosing bronchogenic carcinoma and other endobronchial neoplasms (Figure 20-5); it is far less effective at detecting other causes of hemoptysis. The most common non-neoplastic diagnosis made by FOB is acute bronchitis, which is based on the presence of mucosal hyperemia and edema and purulent-appearing secretions. As previously discussed, however, these findings are nonspecific and often are unrelated to the actual cause of hemoptysis. When this diagnosis is excluded, a specific, non-neoplastic cause of hemoptysis is found by FOB in less than 10% of cases.

Figure 20-5 Bronchogenic carcinoma visualized through the fiberoptic bronchoscope. The tumor occludes the left upper lobe and is actively bleeding.

Computed Tomography Versus Bronchoscopy

Six studies have compared the sensitivity rates for CT and FOB for detection of bronchogenic carcinomas and other lung neoplasms in patients with hemoptysis. Five of these studies were published between 1990 and 1999 and used nonhelical, single-detector scanners. In most cases, images were acquired at 10-mm intervals, although additional sections usually were obtained through the central airways. A total of 296 patients were included in these studies, in whom the chest radiograph typically was either normal-appearing or nonlocalizing, and 83 were diagnosed with a pulmonary malignancy. CT demonstrated all but 1 tumor found by FOB and identified 12 cancers that were not detected by FOB.

Since that time, of course, a dramatic evolution has occurred in CT technology. The advent of continuous, helical scanning using multiple detectors has led to tremendous improvement in image quality and resolution. In a recently published retrospective comparison of FOB and CT in 270 patients with hemoptysis and a normal-appearing chest radiograph, FOB and CT detected 14 and 25 of 26 lung malignancies, respectively. The 1 tumor missed by CT was found by FOB.

These studies clearly show that CT is much better than FOB at detecting and excluding malignancy. However, because CT can still miss small, endobronchial lesions and FOB can establish a definitive tissue diagnosis, both of these tests are recommended, to provide important complementary information in the evaluation of patients with hemoptysis.

Still more recent advances in CT technology have further improved imaging of the airways. Manipulation of CT data allows the creation of two-dimensional multiplanar images as well as three-dimensional internal views of the airways; the relevant technology often is referred to as “virtual bronchoscopy.” Although these advancements would be expected to make CT an even more powerful tool, no studies have evaluated their efficacy in patients with hemoptysis.

Diagnostic Algorithm

A suggested approach to the patient with hemoptysis that is based on the preceding information is shown in Figure 20-6. If the initial evaluation yields a firm diagnosis, such as bacterial pneumonia or iatrogenic or traumatic lung injury, appropriate therapy is instituted. Alternatively, the initial evaluation may suggest a cause of hemoptysis that requires investigation using one or more specific tests. For example, an echocardiogram may confirm the presence of left ventricular failure or mitral stenosis, pulmonary embolism may be diagnosed by means of a CT angiogram, and sputum cultures may be diagnostic in patients with tuberculosis. In all other patients, CT is the most appropriate next step in the diagnostic evaluation. As discussed previously, CT often identifies an unsuspected cause of hemoptysis, even in patients whose chest radiograph is normal in appearance or nonlocalizing, and may provide important information in patients who already have a presumptive diagnosis. For example, in patients with suspected bronchogenic carcinoma, CT adds vital information for staging and also provides a “road map” for bronchoscopy by defining the exact location of a parenchymal mass and enlarged mediastinal lymph nodes.

Figure 20-6 Diagnostic algorithm for patients presenting with hemoptysis.

If CT suggests a disorder that is amenable to bronchoscopic diagnosis, such as neoplasm or infection, FOB is performed next in the diagnostic evaluation. Additional studies such as mediastinoscopy or surgical lung biopsy may be required if FOB is nondiagnostic. When CT either yields normal findings or demonstrates another cause of hemoptysis, such as bronchiectasis, the role of FOB is less clearly defined. The absence of an endobronchial lesion on the CT scan is associated with a low risk of malignancy, and available data suggest that FOB may be safely omitted in never-smokers younger than 40 years. Because CT occasionally fails to detect small endobronchial lesions, however, FOB should be performed in all other patients.

Massive Hemoptysis

No generally accepted definition of massive hemoptysis has emerged, although the most commonly used criteria require the expectoration of between 200 and 600 mL of blood over 24 hours. Any definition based on the amount of expectorated blood is, of course, arbitrary, especially because the volume often is difficult to quantify.

From a clinical standpoint, it is more appropriate to define massive hemoptysis simply as bleeding that impairs ventilation and gas exchange. Because this effect depends not only on the volume of expectorated blood but also on the rate of bleeding, the ability of the patient to clear blood from the airways, and the extent and severity of any underlying lung disease, it is evident that the amount of bleeding needed to be considered “massive” will vary significantly from patient to patient.

Massive hemoptysis is relatively uncommon and occurs in fewer than 5% of patients with lower respiratory tract bleeding. Although any of the disorders listed in Box 20-1 may potentially give rise to life-threatening hemorrhage, massive hemoptysis most commonly is caused by bronchiectasis, bronchogenic carcinoma, mycetoma, lung abscess, and tuberculosis (active or inactive). Overall, the risk of death from massive hemoptysis is approximately 20%, although reported mortality rates vary widely, ranging between 0 and 75%.

The lungs have a dual blood supply. The pulmonary circulation has low pressure and resistance, carries the entire output of the right ventricle, and is responsible for the exchange of oxygen and carbon dioxide. The bronchial arteries are very small systemic arteries with high pressure and resistance. They usually arise from the aorta and normally contribute less than 1% of the blood flow entering the lung vasculature. The bronchial circulation provides oxygen and nutrients to the conducting airways, the pulmonary arteries and veins, the visceral pleura, and the esophagus. Chronic inflammatory disease or neoplasms within the lungs causes enlargement and proliferation of the bronchial arteries as well as recruitment of collateral vessels from extrapulmonary systemic arteries. These vessels tend to have thin walls and are more likely to rupture when exposed to systemic arterial pressure. Massive hemoptysis originates from a bronchial artery in about 90% of cases and from a nonbronchial systemic artery or a pulmonary artery in the remainder. Occasionally, bleeding comes from more than one type of artery.

Because of its associated morbidity and mortality, massive hemoptysis constitutes a respiratory emergency and necessitates rapid evaluation and therapy. Unlike in patients with small amounts of bleeding, in whom the emphasis is on determining the underlying cause, in patients with massive hemoptysis the goals are to maintain a patent airway and to localize and control the bleeding. Patients should be closely monitored in an intensive care unit, and intubation and mechanical ventilation are indicated if ventilation and gas exchange become sufficiently compromised.

Bronchoscopy should be performed immediately in an effort to identify the cause of bleeding or at least to localize the bleeding to a specific segment or lobe. Either rigid bronchoscopy or FOB may be used, depending largely on the clinical circumstances. Rigid bronchoscopy, with its large lumen, affords excellent airway control and suctioning capability, and is ideally suited to management of patients with very brisk bleeding. Disadvantages include poor visualization of the segmental and lobar bronchi and the need for general anesthesia. In most patients, FOB is the procedure of choice, because it can be performed rapidly and allows visualization of airways down to the subsegmental level. All patients with massive hemoptysis should be intubated before FOB. Implementation of this precaution optimizes airway control, allows effective suctioning should the rate of bleeding increase, and permits the bronchoscope to be easily removed and reinserted if the suction channel becomes occluded.

CT also is very effective in identifying both the cause and the site of bleeding in patients experiencing massive hemoptysis. In addition, CT angiography usually can provide images of the bronchial circulation that increase the ease and effectiveness of bronchial artery embolization (see later on) and identify patients who have a nonbronchial systemic arterial source of bleeding.

Localization of the bleeding site is important for two reasons. First, it provides a guide for embolization therapy to control ongoing hemorrhage (see further on). Second, in the setting of persistent, severe hemoptysis, it allows isolation of the bleeding site to prevent the spread of blood throughout the tracheobronchial tree. This potentially lifesaving intervention is performed by placing a bronchoscopically guided balloon catheter in a segmental or lobar airway. When bleeding can be localized only to one lung, a larger balloon may be inflated in a main bronchus, or the fiberoptic bronchoscope can be used to selectively intubate and ventilate the nonbleeding lung.

Once the bleeding site has been localized and a stable airway has been achieved, ongoing hemorrhage must be controlled. Guided by the results of bronchoscopy and CT angiography, bronchial or other intrathoracic arteries may be visualized by use of selective arteriography and occluded with embolized, nonabsorbable material. Arterial embolization is successful in rapidly controlling hemorrhage in 73% to 98% of attempted procedures. Emergent surgical resection is accompanied by a mortality rate that approaches 30% and usually is reserved for patients in whom arterial embolization is unsuccessful.

Once bleeding has resolved, either spontaneously or after embolization therapy, its cause, if not already identified, must be determined (see Figure 20-6). Specific treatment, such as antibacterial or antituberculosis therapy, often prevents further episodes of hemoptysis. When effective therapy is not available, recurrent and often life-threatening hemorrhage occurs with a frequency as high as 50%. Recurrent bleeding is especially likely in patients with bronchiectasis, a mycetoma, or pulmonary malignancy, and surgical resection should be considered in these cases. For patients who are not surgical candidates, repeat embolization therapy often is successful if bleeding recurs.

Summary

Hemoptysis is a commonly encountered sign of extreme clinical importance in that it often signifies the presence of an underlying and sometimes unrecognized lung disease. Adherence to a thorough and orderly diagnostic approach to determine its cause is essential. Massive hemoptysis constitutes a life-threatening emergency, and therapy must focus on maintaining a patent airway and on identification and occlusion of the bleeding artery by means of selective arteriography and embolization.