Show Summary Details
Page of

Therapeutic approach in haemoptysis 

Therapeutic approach in haemoptysis
Therapeutic approach in haemoptysis

Francesco Blasi

and Paolo Tarsia

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2022. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 17 May 2022

Key points

  • The diagnostic work-up involves the complementary use of CT scan, bronchoscopy, and pulmonary angiography.

  • Current multidetector CT angiography is currently considered superior and less time-consuming than traditional pulmonary angiography in identifying the artery causing bleeding.

  • The priority in management of massive haemoptysis involves prompt resuscitation and airway protection with patient stabilization, followed by medical and endoscopic temporary procedures.

  • Super selective bronchial artery embolization (BAE) is now considered the treatment of choice in most patients with massive haemoptysis.

  • In cases of recurrent haemoptysis following BAE, particularly in particular conditions such as mycetoma, surgery may still be considered a desirable option, provided the patient is sufficiently fit.

Diagnostic work-up

The aim of diagnostic studies in patients with haemoptysis is two-fold—locate the source of bleeding, and identify the underlying cause. In investigating a patient with haemoptysis, it is important to ascertain that the blood arises from the lower airways and exclude haematemesis from the gastrointestinal tract or epistaxis from the nasopharynx. Alkaline pH (using pH indicator paper), foaminess, and the presence of pus are useful bedside signs that may indicate the lungs are the source of bleeding prior to diagnostic investigation.

Complete past history regarding respiratory and cardiac diseases should be recorded. A history of co-existent haematuria may suggest the presence of diseases such as Wegener’s granulomatosis and Goodpasture’s syndrome. Current medications, particularly anticoagulant therapy should also be noted. The patient should undergo detailed respiratory and cardiac examination, and nutritional status should be assessed. Additional important signs include finger clubbing, presence of palpable lymph nodes, and signs of skin or mucosal bleeding.

Blood tests should evaluate clotting profile, platelet count, and haemoglobin levels. Repeat testing may be necessary to better quantify the entity of blood loss. Blood gas analysis is essential to assess gas exchange. Sputum testing should be performed for bacteria, mycobacteria and fungi, and malignant cells, particularly in long-term smokers.

Chest X-ray

In most cases, this is the first radiographic investigation performed, and may be informative regarding a number of conditions involving the lung parenchyma, pulmonary vasculature, or the heart. Both a posterior–anterior and a lateral film should be obtained. Pulmonary masses, cavitary lesions, pneumonia, chronic pulmonary disease, or atelectasis may be visualized. Even though the true nature of the disease may not be immediately apparent, a chest radiograph may identify and localize the site of bleeding to one lung or a single lobe in 33–82% of cases [1,2], allowing more focused additional testing. In approximately 20–40% of cases, initial chest X-ray may be normal or non-informative (showing chronic or unchanged abnormalities) [3]‌. A normal chest radiograph is not sufficient to stop investigations for identifying the condition underlying the haemoptysis, as additional testing such as CT scan may reveal conditions such as pulmonary emboli, pulmonary hypertension, bronchiectasis, or initial interstitial disease, not detectable by a plain chest radiograph. Furthermore, it has been reported that between 5 and 6% of patients with haemoptysis, and a normal chest radiograph, a lung cancer is eventually identified [4].

Chest CT scan

Chest CT scan may allow correct localization of the bleeding site in 65–100% of cases [1,5,6]. Several studies demonstrate that CT is at least equal if not superior to bronchoscopy in identifying the cause of haemoptysis [5,6]. The site of haemorrhage may be localized due to detection of liquefied material within bronchi or ground glass infiltrates in the lung parenchyma. Recently, multidetector computed tomography (MDCT) techniques allow scans of the whole thorax in a single breath hold (little more than 10 seconds), with slice thickness approaching 1 mm. Contrast-enhanced MDCT with both two- and three-dimensional volumetric imaging allow high resolution angiographic studies [7]‌. MDCT angiography is particularly useful to interventional radiologists prior to planning bronchial arterial embolization.

In a study comparing traditional invasive angiography with CT angiography, the latter allowed more precise demonstration of both bronchial and non-bronchial systemic arteries [8]‌. Multidetector CT has also proven effective in identifying haemoptysis deriving from the pulmonary artery circulation [9].

CT scans also present some limitations. In severe unstable patients or in cases with active bleeding that requires airway management, shifting the patient to the radiology department may be problematic and bedside bronchoscopy is often preferred in this context.


This procedure may be of help in a number of ways following or during haemoptysis events. It may allow identification of the lung or lobe causing the bleeding, identify the underlying cause, it may help clear the airways from blood clots favouring gas exchange, and finally it may be a means by which therapeutic manoeuvres to stop the bleeding are put in place. It also allows airway sampling, thus obtaining specimens for identification of bacterial, fungal, mycobacterial, or tumoural causes of haemoptysis. Bronchoscopy allows identification of bleeding up to the fifth or sixth generation of bronchi.

Optimal timing for bronchoscopy is still a matter of debate. Bronchoscopy is often performed ‘early’ (within the first 12–18 hours) due to the fact that bleeding may increase with time rendering later visualization of the airways less effective. This may be more pertinent for clinically stable patients or when bleeding has stopped. On the other hand, in patients with massive haemoptysis it is now possible to perform MDCT angiography or traditional invasive angiography with immediate embolization of culprit bronchial arteries within a very short time span. In these patients, bronchoscopy may be performed ‘late’ as an aid to identify the underlying cause of haemoptysis, and allowing microbiological and cytological analysis. There is, as yet, no definitive proof that ‘early’ bronchoscopy is associated with better outcomes.

Rigid bronchoscopy was the procedure of choice in the past. The procedure allows aspiration of large quantities of blood, use of therapeutic manoeuvres, while ensuring adequate ventilation [10]. One drawback of rigid bronchoscopy is the more limited capacity to explore the peripheral bronchial tree and the upper lobes compared with flexible bronchoscopy. Flexible bronchoscopy has now become the most common procedure in patients with haemoptysis, as it may be performed at the patient’s bedside and allows visualization of segmental and subsegmental bronchi [11]. Due to the limited size of its operational channel, flexible bronchoscopy has a more limited suctioning capacity compared with rigid bronchoscopy. Fibre optic bronchoscopy identifies the site of bleeding in 73–93% of cases [11], with a lower diagnostic yield recorded in cases of mild-to-moderate haemoptysis.

Within the diagnostic framework of haemoptysis, bronchoscopy, and CT scan should not be considered as alternative procedures, but rather as complementary, each allowing specific information regarding site and cause identification.

Pulmonary angiography

Diagnostic pulmonary angiography was traditionally considered the procedure of choice to identify the source of bleeding. Moreover, it allows bronchial artery embolization during the same procedure. However, given the variability of the bronchial artery supply, complete angiographic investigation of orthoptic and ectopic bronchial arteries to identify the origin of bleeding may be quite time consuming. If CT, bronchoscopy, or both are performed prior to angiography, a better definition of the site of bleeding allows considerable shortening of procedure time. During angiography, signs that may be used to determine the site of bleeding include vessel size, vascular blush, focal hypervascularization, and presence of vascular shunts.

Over the last decade, it has become apparent that diagnostic angiography is less effective than MDCT angiography in detecting the site of bleeding [8]‌.

Therapeutic approach

Treatment of haemoptysis may vary considerably, ranging from outpatient management to intensive care unit admittance. Choice of optimal management depends on the intensity of bleeding, degree of respiratory compromise, and severity of underlying cardiorespiratory status. Mild haemoptysis may be handled on an outpatient basis with antibiotics and symptomatic treatment, such as antitussives. For more severe cases, hospitalization is required for diagnostic and therapeutic work-up. Blood abnormalities must be corrected with corresponding blood products. Tranexamic acid is an antifibrinolytic agent that acts by inhibiting plasminogen activation. Oral or intravenous formulations of this agent have been employed in patients with mild-to-moderate haemoptysis. Intravenous vasopressors do not seem to have a role in the routine management of haemoptysis due to concerns in patients with co-existing coronary artery disease or hypertension. It has been suggested that bronchodilator use may best be limited in patients with haemoptysis as these drugs may exert vasodilator effects, thus facilitating bleeding recurrence [8]‌.

Important steps in the management of patients with massive haemoptysis include:

  • Resuscitation, airway protection, and patient stabilization as the priority.

  • Subsequent localization of the site of bleeding.

  • Specific interventions to stop the bleeding and prevent recurrence.

Intubation may be required with a large bore tube (preferably size 8 or more) to facilitate airway suctioning. A double-lumen endotracheal tube may be used when diseased side isolation is required in order to protect the unaffected lung. The left-sided tube is easier to place compared with the right-sided tube. The latter carries a higher risk of obstructing the right upper lobe bronchus, given that it originates rather proximally along the right main bronchus. Initial management also includes volume resuscitation, and correction of underlying coagulation disorders. If the site of bleeding includes only one lung, the affected side should be kept dependent in lateral decubitus in order to prevent aspiration to the contralateral side. The true clinical validity of this theoretical consideration has not been test in controlled trials.

Bronchoscopic treatment

As discussed earlier, bronchoscopy may be of use in a number of ways in patients with haemoptysis, allowing site of bleeding identification, sampling of the lower airways, airway clearing, and institution of therapeutic manoeuvres. The relative advantages and disadvantages of rigid versus fibre optic bronchoscopy that have been discussed regarding the diagnostic use of bronchoscopy, apply to the therapeutic aspects of the technique also.

Instillation of different types of substances has been used in the treatment of patients with haemoptysis. Cold saline lavage may be performed with 50-mL aliquots of refrigerated (4°C) saline, instilling roughly 500 mL overall volume. This technique may be effective in stopping bleeding [11]. A rigid bronchoscope is generally preferred due to better suction capacity compared with fibre optic bronchoscopy.

Topical vasoconstrictive agents have been tested, mainly in cases of bronchoscopy-induced airway bleeding. Adrenaline (1:20,000) [11], or antidiuretic hormone derivatives, such as terlipressin (0.5 mg) and ornipressin (5 IU) have been used [12], but their efficacy in massive haemoptysis is debated since the drugs are easily diluted and washed away. Likewise, endoscopic instillation of fibrinogen-thrombin or thrombin has been evaluated, but efficacy has not been convincingly demonstrated [13]. Endobronchial application of tranexamic acid (500–1000 mg) may be used in mild-to-moderate cases. Temporary endobronchial tamponade with a balloon catheter is a commonly employed procedure. 4 Fr 80-cm long Fogarty catheters can be inserted through a flexible bronchoscope and ensure emergency control of bleeding while awaiting bronchial artery embolization or surgery [14]. Alternatively, Freitag et al. [15] developed a 6 Fr 170-cm long double lumen balloon catheter. The second lumen facilitates instillation of cold saline or topical agents. A detachable valve at the proximal end of the catheter allows easy removal of the bronchoscope with minimal risk of balloon displacement. The prolonged use of balloon catheter tamponade is discouraged as ischaemic mucosal injury and post-obstructive pneumonia may ensue.

A number of endobronchial materials or procedures have been used to facilitate temporary bleeding interruption, such as airway stents, silicone spigots, oxidized regenerated cellulose mesh, sealing glues, Nd-YAG laser, cryotherapy, and brachytherapy, particularly in the context of haemoptysis caused by endobronchial carcinoma [11].

Bronchoscopy treatment strategies may allow temporary control of bleeding, buying time while patients’ conditions are stabilized. In some instances, the procedures may provide long-lasting haemostatic effects, but in many cases of massive haemoptysis provides temporary relief while awaiting more definitive treatment, such as bronchial artery embolization or surgery.

Bronchial artery embolization

Bronchial artery embolization (BAE) is performed during pulmonary angiography by occluding systemic vessels and blocking arterial blood flow to fragile vessels within diseased tissue. Non-selective BAE was associated with a greater rate of inadvertent occlusion of spinal arteries. More recently, microcatheter superselective bronchial artery catheterization has allowed more stable positioning within the bronchial tree and more distal cannulation avoiding spinal artery embolization.

BAE is now considered the procedure of choice in patients with persistent haemoptysis following medical management, massive haemoptysis, recurrence of haemoptysis, in candidates deemed unfit for surgery, or as an emergency procedure to temporarily control bleeding, while awaiting surgery. Embolizing materials include gelfoam, polyvinyl alcohol (PVA) particles, dextran microspheres, or steel coils. PVA particles, available in a variety of particulate sizes (generally measuring between 350–500 μ‎m in diameter) are the most commonly used agents. These non-absorbable particles prevent recanalization. Metal coils tend to occlude more proximal vessels and render subsequent embolization virtually impossible, but are the material of choice in pulmonary artery aneurysms and in pulmonary arteriovenous malformations. With BAE, immediate control of bleeding may be obtained in 70–95% of patients. Recurrence has been reported in 10–30% of cases [16]. However, the procedure may be repeated safely over time. Recurrence may be due to non-optimal embolization of bronchial arteries, collateral vessel formation, origin of the bleeding from the pulmonary artery circulation, and presence of non-bronchial systemic arteries. Early recurrence (within weeks) of bleeding is generally caused by incomplete embolization, whereas late rebleeding is more likely due to recanalization of embolized vessels or collateral circulation due to progression of the underlying disease condition. Some lung cavitary conditions, such as mycetoma are more prone to recurrence than others and are ultimately best managed by surgery. Persistent recurrence may also be associated with within bleeding arising from the pulmonary artery circulation.

Common complications following BAE include fever, dysphagia (present in 1–18% of cases), and chest or back pain (in 24–91% of cases) [17]. These side effects are generally transitory and are probably the result of the occlusion of vessels supplying the oesophagus, and diaphragmatic and pleural visceral pleura. The most feared complication is paraplegia derived from occlusion of spinal arteries that arise from bronchial or intercostal arteries. Use of superselective microcatheters has currently reduce the rate of this complication to <1%. In patients with massive haemoptysis the death rate following BAE is between 7 and 18% [18].


Prior to BAE, surgery was considered to be the most effective method of controlling massive haemoptysis. Surgical mortality may be as high as 40% in the emergency setting, with additional morbidity in another 25–50% [19]. Mortality is considerably greater in emergency situations than in elective surgery. Moreover, most case series are now over 20 years old, and mortality has probably dropped considerably over the last decade due to improved surgical techniques and expertise.

Surgery is still considered the strategy of choice in conditions known to exhibit poor response to BAE, such as mycetoma. Additional situations where surgery is still indicated are complex arteriovenous malformations, iatrogenic pulmonary artery rupture, localized bronchiectasis and tuberculosis scars. Radiation therapy is a further option in patients with mycetoma or vascular tumours with massive recurrent haemoptysis [20]. Its effects are mediated through necrosis of feeding blood vessels and vascular thrombosis due to perivascular oedema.

Haemoptysis treatment algorithm

Haemoptysis, particularly massive haemoptysis, may present as a clinical emergency. Prompt patient stabilization, diagnostic testing, and therapeutic management are best obtained through multidisciplinary activation of intensivists, pulmonologists, radiologists, and thoracic surgeons. Fig. 127.1 presents a treatment algorithm that summarizes diagnostic and therapeutic options available for patients with massive haemoptysis.

Fig. 127.1 Treatment algorithm for the management of haemoptysis.

Fig. 127.1 Treatment algorithm for the management of haemoptysis.


1. Khalil A, Soussan M, Mangiapan G, Fartoukh M, Parrot A, and Carette MF. (2007). Utility of high-resolution chest CT scan in the emergency management of hemoptysis in the intensive care unit: severity, localization and aetiology. British Journal of Radiology, 80, 21–5.Find this resource:

2. Ong TH and Eng P. (2003). Massive hemoptysis requiring intensive care. Intensive Care Medicine, 29, 317–20.Find this resource:

3. Flower CDR and Jackson JE. (1996). The role of radiology in the investigation and management of patients with hemoptysis. Clinical Radiology, 51, 391–400.Find this resource:

4. Herth F, Ernst A and Becker HD. (2001). Long-term outcome and lung cancer incidence in patients with hemoptysis of unknown origin. Chest, 120, 1592–4.Find this resource:

5. Revel MP, Fournier LS, Hennebicque AS, et al. (2002). Can CT replace bronchoscopy in the detection of the site and cause of bleeding in patients with large or massive hemoptysis? American Journal of Roentgenology, 179, 1217–24.Find this resource:

6. Hsiao EI, Kirsch CM, Kagawa FT, et al. (2001). Utility of fibreoptic bronchoscopy before bronchial artery embolization for massive hemoptysis. American Journal of Roentgenology, 177, 861–7.Find this resource:

7. Bruzzi JF, Rémy-Jardin M, Delhaye D, et al. (2006). Multi-detector row CT of hemoptysis. Radiographics, 26, 3–22.Find this resource:

8. Remy-Jardin M, Bouaziz N, Dumont P, et al (2004). Bronchial and nonbronchial systemic arteries at multi-detector row CT angiography: comparison with conventional angiography. Radiology, 233,741–9.Find this resource:

9. Khalil A, Parrot A, Nedelcu C, Fartoukh M, Marsault C, and Carette MF. (2008). Severe hemoptysis of pulmonary arterial origin: signs and role of multidetector row CT. Chest, 133, 212–19.Find this resource:

10. Karmy-Jones R, Cuschieri J, and Vallieres E. (2001). Role of bronchoscopy in massive haemoptysis. Chest Surgery Clinics of North America, 11, 873–906.Find this resource:

11. Sakr L and Dutau H. (2010). Massive haemoptysis an update on the role of bronchoscopy in diagnosis and management. Respiration, 80, 38–58.Find this resource:

12. Tüller C, Tüller D, Tamm M, and Brutsche MH. (2004). Hemodynamic effects of endobronchial application of ornipressin versus terlipressin. Respiration, 71, 397–401.Find this resource:

13. De Gracia J, Dela Rosa D, Catalan E, Alvarez A, Bravo C, and Morell F. (2003). Use of endoscopic fibrinogen-thrombin in the treatment of severe hemoptysis. Respiratory Medicine, 97, 790–5.Find this resource:

14. Gottlieb LS and Hillberg R. (1975). Endobronchial tamponade therapy for intractable hemoptysis. Chest, 67, 482–3.Find this resource:

15. Freitag L, Telkolf E, Stamatis G, Montag M, and Greschuchna D. (1994). Three years experience with a new balloon catheter for the management of haemoptysis. European Respiratory Journal, 7, 2033–7.Find this resource:

16. Lee EW, Grant JD, Loh CT, and Kee ST. (2008). Bronchial and pulmonary arterial and venous interventions. Seminars in Respiratory and Critical Care Medicine, 29(4), 395–404.Find this resource:

17. Chun JY, Morgan R, and Belli AM. (2010). Radiological management of hemoptysis: a comprehensive review of diagnostic imaging and bronchial arterial embolization. Cardiovascular and Interventional Radiology, 33, 240–50.Find this resource:

18. Yoon W, Kim JK, Kim YH, Chung TW, and Kang HK. (2002). Bronchial and nonbronchial systemic artery embolization for life-threatening hemoptysis: a comprehensive review. Radiographics, 22, 1395–409.Find this resource:

19. Bussieres JS. (2001). Massive hemoptysis. In: P. Slinger (ed.) Principles and Practice of Anesthesia for Thoracic Surgery . New York: Springer Science + Business Media LLCFind this resource:

20. Shneerson JM, Emerson PA, and Philips RH. (1980). Radiotherapy for massive haemoptysis from an aspergilloma. Thorax, 35, 953–4.Find this resource: