Hemoptysis generates great concern for both the patient and the clinician. Although massive hemoptysis occurs infrequently, it must be recognized promptly and dealt with appropriately to minimize the risk of mortality. In this article, we provide an overview of hemoptysis in the adult. We concentrate on the common causes, offer a rational approach to evaluation of the patient, and describe therapeutic options and their appropriate application. OVERVIEW Hemoptysis--defined as the expectoration of blood originating from the lower respiratory tract--derives from the Greek terms, "haima" (blood) and "ptysis" (spitting). The classification of hemoptysis severity is based on volume of expectorated blood in 24 hours. In the literature, 600 mL in 24 hours is a common definition of "massive," although a range of 200 to 1000 mL has been cited.1 Since the actual volume is almost never measured, most chest physicians become concerned when an estimated 200 to 400 mL is expectorated and use this value as a definition of massive hemoptysis. Mild or minimal hemoptysis usually refers to "specks" of blood or a few small clots in sputum, and moderate hemoptysis is everything from "specks" to 200 mLin 24 hours. One reason to classify hemoptysis is to establish prognosis. Reported mortality in massive hemoptysis has ranged from less than 10% with intervention to 85% without intervention.2,3 Perhaps more important than expectorated volume is the rate of bleeding. Crocco and associates4 reported a mortality rate of 71% if 600 mL was expectorated in 4 hours or less, 22% if this occurred in 4 to 16 hours, and 5% if in 16 to 48 hours. Fortunately, massive hemoptysis is infrequent, with an incidence of less than 10%.1,2,5 The assessment of the severity of hemoptysis determines the subsequent evaluation and therapeutic strategy. With massive or moderate hemoptysis, the safest course is to hospitalize the patient and consult a pulmonologist or a thoracic surgeon. Evaluation and treatment occur concomitantly. Patients with mild or minimal hemoptysis can be evaluated as outpatients, and treatment can be instituted when the cause is established. ETIOLOGY The lungs receive blood from 2 relatively independent systems--the high-pressure bronchial and low-pressure pulmonary arterial systems. Most hemoptysis originates from the bronchial arterial circulation (90%), with the remainder from the pulmonary arterial circulation and a variety of other sources.6 The pathogenesis of hemoptysis is variable but frequently involves abnormal vasculature at risk for injury. Common mechanisms include proliferation of fragile vessels (because of chronic infection, inflammation, or malignancies), erosion of blood vessels (by infection or malignancy), antibody or other immunological vascular injury (as in Goodpasture syndrome or systemic lupus erythematosus [SLE]), excessive pulmonary venous distention (mitral stenosis), and arteriovenous malformation. There are many causes of hemoptysis (Table 1). Pulmonary tuberculosis may be the major cause of hemoptysis worldwide. Other common causes include malignancy, bronchiectasis, pneumonia, and bronchitis. Reisz and associates,7 in a review of their institution's data from 1986 to 1995 reported the following causes: bronchitis (26%); malignancy, primarily lung cancer (25%); pneumonia (10%); mycobacterial infections, primarily pulmonary tuberculosis (8%); and idiopathic (7%). Santiago and associates8 published similar data, except that the idiopathic rate in their study was higher (22%). The incidence of the various causes of hemoptysis varies across countries and time periods. Data from several institutions, locations, and time periods (1940 - 1995) are shown in Table 2. Table 3 summarizes the frequently reported causes of massive hemoptysis, which include bronchiectasis, pneumonia, and lung abscess. Moderate or massive hemoptysis may be more common with a necrotizing pneumonia, such as that caused by Staphylococcus aureus or Klebsiella pneumoniae, than with community-acquired pneumonia. In his retrospective study of 208 patients, Hirshberg and associates9 reported that bronchiectasis, lung cancer, pneumonia/lung infection, and hemorrhagic diathesis were the most common causes of massive hemoptysis. Lung cancer accounted for 20% to 45% of cases in several studies.7,8 Hemoptysis is rare in children. Godfrey10 reviewed 2148 pediatric bronchoscopies done in a tertiary referral center and found that only 17 were for hemoptysis. Tracheobronchitis, pneumonia, lung abscess, foreign body, and tumors (adenoma, carcinoid) are cited as the common causes of hemoptysis in children.11 In another tertiary center, cystic fibrosis accounted for 65% and congenital heart disease for 15% of cases.12 Infections are the main cause of hemoptysis in patients with HIV infection. Nelson and Forman13 reported that 32 (63%) of 51 cases of hemoptysis in HIV-positive patients were the result of infection, and 11 (22%) were idiopathic. Only 2 patients had massive hemoptysis. Patients with idiopathic hemoptysis generally have a good prognosis. Adelman and colleagues14 studied 67 patients who had nonmassive hemoptysis and normal findings on chest radiography and bronchoscopy; in 90%, the hemoptysis stopped within 6 months; in 1 patient, lung cancer was diagnosed about 20 months later. DIAGNOSIS History and physical examination Although the history and physical examination findings rarely are pathognomonic of hemoptysis, they can confirm hemoptysis and provide valuable clues to the cause. The initial objectives are to differentiate hemoptysis from epistaxis and hematemesis and then to establish the severity. The appearance of specks of blood a few days after the acute onset of a productive cough, congestion, sore throat, and low-grade fever suggests bronchitis. Malignancy is a primary concern in patients who present with hemoptysis, especially those with risk factors such as age over 50 years, male sex, and a smoking history of more than 40 pack-years.15 Weight loss, fever, and night sweats suggest infection, such as pulmonary tuberculosis or lung abscess. Patients with hemoptysis caused by pulmonary embolism are frequently dyspneic. In these patients, hemoptysis is generally a late phenomenon; the hemoptysis represents pulmonary infarction, which usually means there is a large blood clot. Hemoptysis also is a late presentation of both pulmonary arterial hypertension and mitral stenosis. In the latter, dyspnea usually precedes the hemoptysis. A history of rheumatic fever and progressive exertional dyspnea in a patient with sudden hemoptysis or the pink, frothy sputum associated with pulmonary edema points to mitral stenosis. Concomitant hematuria suggests a vasculitis or immunologically mediated disease, such as Wegener granulomatosis, SLE, or Goodpasture syndrome. A cyclical pattern associated with menses can occur with a rare condition, thoracic endometriosis. The physical examination findings occasionally suggest the diagnosis, but more reliably indicate the level of the patient's distress, since blood in the alveoli often leads to hypoxemia and respiratory embarrassment. With approximately 400 mL of blood in the alveoli, significant hypoxemia, asphyxiation, and death can occur.1 Rarely, shock due to exsanguination causes death. Nasal ulcerations suggest Wegener granulomatosis, and oral mucosal and cutaneous telangiectases suggest hereditary hemorrhagic telangiectasia and arteriovenous malformation. An opening snap and a diastolic murmur may suggest mitral stenosis, whereas an accentuated second heart sound in the pulmonic valve area suggests pulmonary hypertension. Findings on chest auscultation are variable. Localized wheeze caused by bronchial obstruction can occur with an endobronchial lesion, such as lung cancer. In patients with diffuse alveolar hemorrhage or massive hemoptysis, there may be rhonchi and crackles or fairly normal auscultatory findings. Clubbing can occur with lung cancer, bronchiectasis, or lung abscess. Radiology For patients with hemoptysis, a chest radiograph is important for both diagnostic and therapeutic purposes. The findings can suggest the diagnosis or localize the side of bleeding in most cases (Figure 1). However, the findings may be normal or nonlocalizing in 20% to 46% of patients.6 About 5% to 6% of patients with hemoptysis who have normal chest radiographic findings may have a lung cancer.16 Chest CT scanning provides more detailed images than radiography and can identify sources of bleeding that are not apparent on radiographs, such as bronchiectasis, pulmonary embolism, arteriovenous malformation, aspergilloma, and lung cancer.17-19 To evaluate for pulmonary embolism or vascular malformation, intravenous contrast is needed. The role of chest CT in hemoptysis has not yet been fully defined. In one study, CT was thought to identify the likely source of bleeding in 50% of patients who had normal findings on chest radiography and bronchoscopy, supporting a role in this subgroup of patients.18 Compared with bronchoscopy, CT generally has higher yield in identifying the likely bleeding source.18 CT can be used to direct biopsies and improve the yield of bronchoscopy. Hirshberg and coworkers9 reported that the combination of chest CT and bronchoscopy contributed to the final diagnosis in 93% of cases, compared with 42% with bronchoscopy alone, 50% with chest radiography alone, and 67% with chest CT alone. Even if the diagnosis cannot be made, the combination of chest CT and bronchoscopy often localizes the site of bleeding, which is important in massive and moderate hemoptysis, since this guides either radiological or surgical intervention. These techniques should be viewed as complementary examinations. Laboratory Certain laboratory studies assist in the diagnosis and management of hemoptysis. These include measurement of hemoglobin and hematocrit to assess blood loss and platelet count, international normalized ratio, activated partial thromboplastin time, and creatinine level to assess coagulation status. These tests should be done in patients with persistent mild or minimal, moderate, or massive hemoptysis. However, if bronchitis is the working diagnosis, the laboratory tests may be delayed. Pulse oximetry or arterial blood gas analysis to document oxygenation is indicated for massive and moderate hemoptysis. Other tests are based on the results of the history, physical examination, and radiological assessment. Sputum stains and cultures and, in some cases, blood cultures are reasonable when pneumonia, cavitary disease, or lung abscess is suspected. When a vasculitis or an immunologically mediated vascular disease is suspected, measuring antineutrophil cytoplasmic antibodies (ANCA), anti-glomerular basement membrane antibodies, antinuclear antibodies, anti-double-stranded DNA (dsDNA) antibodies, and complement levels is prudent. In most cases of Wegener granulomatosis, the c-ANCA is positive. Anti-basement membrane antibody is usually positive in Goodpasture syndrome. The presence of antinuclear antibodies with anti-dsDNA antibodies supports the diagnosis of SLE.Some clinicians find that the erythrocyte sedimentation rate is a useful and rapid screening test for vasculitis in the urgent setting. In addition, if hematuria coexists, a urinalysis is warranted. If valvular heart disease is suspected, an echocardiogram is appropriate. Bronchoscopy When performed within 48 hours of hemoptysis, bronchoscopy can localize the bleeding in 34% to 91% of cases, compared with 11% to 52% when performed later, but this may not alter therapy.20 For patients with massive hemoptysis, most chest physicians advocate early bronchoscopy because it may provide a diagnosis, localize the site, or allow bronchoscopic intervention to control the bleeding.20,21 In nonmassive hemoptysis, the need for bronchoscopy should be determined on a case-by-case basis. Bronchoscopy is reasonable in patients with risk factors for lung cancer, nonmassive hemoptysis, and normal or nonlocalizing findings on a chest radiograph. In contrast, when bronchitis is the likely cause of hemoptysis, bronchoscopy does not need to be performed unless the problem does not respond to conservative treatment. Most chest physicians use flexible instead of rigid bronchoscopy because of training and availability. Rigid bronchoscopy requires a general anesthetic and is usually performed in the operating room. Fewer physicians are trained in this technique. However, for massive hemoptysis, some chest physicians prefer rigid bronchoscopy, since it provides for better control of the airway, has more efficient suction capability, and allows use of some interventions that are unavailable with flexible bronchoscopy. This decision is up to the treating physician. In patients with less severe hemoptysis, flexible bronchoscopy is the appropriate initial choice. Figure 2 outlines an algorithm for evaluating hemoptysis. TREATMENT For patients with mild or minimal hemoptysis, treatment is directed at the specific cause, such as antimicrobial agents for infection, immunosuppressive agents for vasculitis, or mitral valve repair or replacement for mitral stenosis. Nonspecific therapy includes antitussive agents to minimize cough-induced irritation that may lead to bleed- ing. Idiopathic hemoptysis often resolves spontaneously. Patients who have massive and moderate hemoptysis should be hospitalized, since evaluation and treatment should occur concomitantly. Hypoxemia in the presence of any hemoptysis also warrants hospital admission.The initial goals are to maintain a protected airway, adequate oxygenation and ventilation, hemodynamic stability, and normalized coagulation parameters. If hemoptysis is massive, intubation is recommended, and the patient is placed in the lateral decubitus position, with the "bleeding lung" in the "down" position, if feasible. This optimizes ventilation and perfusion of the unaffected lung. At the time of bronchoscopy, either a balloon or a single- or a double-lumen endotracheal tube may be placed such that the bronchus leading to the bleeding lung is occluded and the unaffected lung is ventilated safely. During bronchoscopy, interventions to control the bleeding may include instilled epinephrine or thrombin, chilled saline lavage (requires rigid bronchoscopy), cryotherapy, laser, electrocautery, or argon plasma coagulation (Figure 3). These are temporizing measures in most cases. When bronchoscopic treatments are insufficient and the bleeding is unilateral, angiography is the next step. Since the bronchial arterial circulation is the likely source of bleeding, this system is assessed initially. The interventional radiologist attempts to locate the vessel(s) responsible for the bleeding and to embolize the vessel(s), using microparticles, absorbable gelatin sponges, or microcoils (Figure 4). Successful embolization occurs in 90% of cases, and bleeding is controlled in 66% to 90% over the first month; however, recurrent hemoptysis (up to 12 months later) has been reported in 20% to 40% of patients.19 Embolization of a bronchial artery is not without complications. Spinal cord injury (in fewer than 1% of patients) resulting from accidental occlusion of spinal or intercostal arteries is possible. Oth- er reported complications include esophageal ulceration, stroke, and bronchial infarction. Localizing the side of bleeding before angiography can minimize the risk by directing the angiographer to the affected side. If the bronchial artery is not the culprit, pulmonary angiography can be undertaken. When successful, embolization may be the definitive therapy. In some cases, it allows for a more thorough evaluation of the underlying cause and for selection of optimal therapy. Embolization can be repeated if needed. Emergent surgery for massive hemoptysis carries a mortality rate of 7% to 50%.5 In this setting, the patient's comorbidities, which might determine whether the patient is a candidate for surgery, are often unknown. In addition, a site or source of bleeding amenable to resection must be identified. Ideally, the hemoptysis can be controlled by the measures discussed above, providing time to determine whether surgery is the best therapy and whether the patient can tolerate surgery. A patient who has diffuse bleeding from both lungs is seldom a candidate for either radiological or surgical intervention. When pulmonary hemorrhage is suspected, as in some cases of SLE or of graft versus host disease after bone mar- row transplant, immunosuppressive therapy (primarily high-dose corticosteroids) may be considered.22 References: 1. Corder R. Hemoptysis. Emerg Med Clin North Am . 2003;21:421-435. 2. Thompson AB, Teschler H, Rennard SI. Pathogenesis, evaluation, and therapy for massive hemoptysis. Clin Chest Med . 1992;13:69-82. 3. Winter S, Ingbar D. Massive hemoptysis: pathogenesis and management. J Intensive Care Med . 1988;3:171-187. 4. Crocco JA, Rooney JJ, Fankushen DS, et al. Massive hemoptysis. Arch Intern Med . 1968;121: 495-498. 5. Lenner R, Schilero GJ, Lesser M. Hemoptysis: diagnosis and management. Compr Ther. 2002; 28:7-14. 6. Jean-Baptiste E. Clinical assessment and management of massive hemoptysis. Crit Care Med . 2000;28:1642-1647. 7. Reisz G, Stevens D, Boutwell C, Nair V. The cause of hemoptysis revisited. A review of the etiologies of hemoptysis between 1986 and 1995. Mo Med . 1997;94:633-635. 8. Santiago S, Tobias J, Williams AJ. A reappraisal of the causes of hemoptysis. Arch Intern Med . 1991;151:2449-2451. 9. Hirshberg B, Biran I, Glazer M, Kramer MR. Hemoptysis: etiology, evaluation, and outcome in a tertiary referral hospital. Chest . 1997;112:440-444. 10. Godfrey S. Pulmonary hemorrhage/hemoptysis in children. Pediatr Pulmonol . 2004;37:476-484. 11. Pianosi P, Al-sadoon H. Hemoptysis in children. Pediatr Rev . 1996;17:344-348. 12. Coss-Bu JA, Sachdeva RC, Bricker JT, et al. Hemoptysis: a 10-year retrospective study. Pediatrics . 1997;100:7-10. 13. Nelson JE, Forman M. Hemoptysis in HIV-infected patients. Chest . 1996;110:737-743. 14. Adelman M, Haponik EF, Bleecker ER, Britt EJ. Cryptogenic hemoptysis. Clinical features, bronchoscopic findings, and natural history in 67 patients. Ann Intern Med . 1985;102:829-834. 15. Poe RH, Israel RH, Marin MG, et al. Utility of fiberoptic bronchoscopy in patients with hemoptysis and a nonlocalizing chest roentgenogram. Chest . 1988;93:70-75. 16. Colice GL. Detecting lung cancer as a cause of hemoptysis in patients with a normal chest radiograph: bronchoscopy vs CT. Chest . 1997;111: 877-884. 17. Bruzzi JF, Remy-Jardin M, Delhaye D, et al. Multi-detector row CT of hemoptysis. Radiographics . 2006;26:3-22. 18. Marshall TJ, Flower CD, Jackson JE. The role of radiology in the investigation and management of patients with haemoptysis. Clin Radiol . 1996; 51:391-400. 19. Andersen PE. Imaging and interventional radiological treatment of hemoptysis. Acta Radiol . 2006; 8:780-792. 20. Dweik RA, Stoller JK. Role of bronchoscopy in massive hemoptysis. Clin Chest Med . 1999;20:89-105. 21. Haponik EF, Fein A, Chin R. Managing life-threatening hemoptysis: has anything really changed? Chest . 2000;118:1431-1435. 22. Chao NJ, Duncan SR, Long GD, et al. Corticosteroid therapy for diffuse alveolar hemorrhage in autologous bone marrow transplant recipients. Ann Intern Med . 1991;114:145-146. 23. Ingbar DH. A systematic workup for hemoptysis. Contemp Intern Med . 1989;1:60-70. 24. Abbott OA. The clinical significance of pulmonary hemorrhage: a study of 1316 patients with chest disease. Dis Chest. 1948;14:824-842. 25. Souders CR, Smith AT. The clinical significance of hemoptysis. N Engl J Med. 1952;247:790-793. 26. Moersch HJ. Clinical significance of hemoptysis. JAMA. 1952;148:1461-1465. 27. Johnston H, Reisz G. Changing spectrum of hemoptysis. Underlying causes in 148 patients undergoing diagnostic flexible fiberoptic bronchoscopy. Arch Intern Med. 1989;149:1666-1668. 28. Alaoui AY, Barial M, el Boutahiri A, et al. Clinical characteristics and etiology in hemoptysis in a pneumology service: 291 cases. Rev Mal Respir. 1992;9:295-300. 29. Knott-Craig CJ, Oostuizen JD, Rossouw G, et al. Management and prognosis of massive hemoptysis. Recent experience with 120 patients. J Thorac Cardiovasc Surg. 1993;105:394-397. 30. McGuinness G, Beacher JR, Harkin TJ, et al. Hemoptysis: prospective high-resolution CT/bronchoscopic correlation. Chest. 1994;105:1155-1162.
