Show Summary Details
Page of

Pulmonary haemorrhagic disorders 

Pulmonary haemorrhagic disorders

Pulmonary haemorrhagic disorders

D.J. Hendrick

and G.P. Spickett


July 30, 2015: This chapter has been re-evaluated and remains up-to-date. No changes have been necessary.


Chapter reviewed August 2013—no substantial updates required; one new reference added.

Updated on 27 Feb 2014. The previous version of this content can be found here.
Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2015. 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).

Subscriber: null; date: 30 March 2017


There are many causes of bleeding within the lung and haemoptysis, but the term ‘pulmonary haemorrhagic disorder’ applies only to diffuse bleeding from pulmonary alveolar capillaries. The condition is characterized by haemoptysis (not invariable), breathlessness, diffuse air space shadowing on the chest radiograph, anaemia (normochromic normocytic if acute, iron deficient with chronicity), and elevated carbon monoxide gas transfer (TLco).

Goodpasture’s disease—diffuse pulmonary haemorrhage and glomerulonephritis with linear deposition of antibodies along the glomerular basement membrane (anti-GBM antibodies). Renal failure is a much commoner threat to survival than lung haemorrhage, but in some cases (almost invariably smokers) the latter can be life threatening. Treatment is supportive (artificial ventilation occasionally necessary) and with steroids, other immunosuppressants (cyclophosphamide) and plasmapheresis.

Idiopathic pulmonary haemosiderosis—a rare disorder of unknown cause with recurrent alveolar bleeding. Chest imaging shows the nonspecific appearances of intra-alveolar blood, which usually clears spontaneously over 1 to 3 weeks, but with chronicity diffuse pulmonary fibrosis with honeycombing may supervene. Treatment is supportive.

Other causes—pulmonary haemorrhage may occur with or complicate a wide variety of disorders with vasculitic, immunological, infective, vascular, haemostatic, toxic, or unknown origins.


Bleeding within the lung and subsequent haemoptysis is common in clinical practice and may be the consequence of many unrelated disorders. The term ‘pulmonary haemorrhagic disorder’ is properly applied only to the particular circumstances of bleeding arising diffusely from pulmonary alveolar capillaries. A preferable and more explicit diagnostic term is therefore pulmonary capillary (or alveolar) haemorrhage, which is not a disease entity of itself, merely a clinical feature of several diseases, but it is a defining characteristic of two, Goodpasture’s syndrome and idiopathic pulmonary haemosiderosis.

Although the lung can accommodate only small quantities of blood in the major airways without threatening life from asphyxiation, it can sequester surprisingly large amounts (litres) at alveolar level. This has a curious effect, unique among diffuse parenchymal diseases of the lung and of considerable diagnostic value; the carbon monoxide gas transfer (TLco) is raised significantly above normal. Not only are physiologically useful red cells within the alveolar capillaries able to absorb the inhaled carbon monoxide, but so too are those lost from the circulation into the alveolar spaces.

Pulmonary capillary haemorrhage is thus characterized by haemoptysis, breathlessness, diffuse air space shadowing on the chest radiograph (Fig., anaemia (normochromic normocytic if acute, iron deficient with chronicity), and an elevated TLco (see Chapter 18.3.1). The extravasated red cells are not readily expectorated, although enough generally escape to cause haemoptysis, and so haemosiderin accumulates within alveolar macrophages as the red cells and their debris are engulfed. The diagnosis of pulmonary capillary haemorrhage is largely confirmed when haemosiderin-laden macrophages are identified in sputum (and there is no localized site of bleeding). If sputum is not expectorated or haemoptysis is absent, minimal, or otherwise explained, then bronchoalveolar lavage and/or lung biopsy are often necessary to establish the diagnosis. An alternative approach is CT or MRI, which may alone provide convincing evidence of blood sited diffusely within the alveoli.

Fig. Radiograph showing gross alveolar shadowing following severe pulmonary haemorrhage in a 60-year-old man with systemic vasculitis.

Radiograph showing gross alveolar shadowing following severe pulmonary haemorrhage in a 60-year-old man with systemic vasculitis.

Although diffuse alveolar capillary haemorrhage may characterize or complicate a wide variety of specific diseases or disease settings, the direct effects of the haemorrhage itself are not influenced by the cause, nor are the means by which it can be recognized. There may, nevertheless, be substantial differences at presentation from case to case according to severity and chronicity.

Goodpasture’s syndrome

Goodpasture described a man with pulmonary haemorrhage and renal failure from glomerulonephritis. A number of conditions can cause such a ‘pulmonary–renal syndrome’, the best-characterized of which (although almost certainly not the illness suffered by the patient in the original report) is now termed Goodpasture’s disease. This consists of diffuse pulmonary haemorrhage and glomerulonephritis with linear deposition of antibodies along the glomerular basement membrane (anti-GBM antibodies), 90% of which are directed against the α‎-3 chain of type IV collagen. Increased susceptibility is associated with HLA DRB11 501 and DRB11 502 alleles, while protection is associated with HLA DR1 and DR7.

Clinical features

In general, glomerulonephritis proves to be a much commoner threat to survival than lung haemorrhage, and the diagnosis of Goodpasture’s disease is reached more conveniently from kidney rather than lung biopsy, together with serological testing for anti-GBM antibodies. In some cases, however, lung disease dominates the clinical picture. Most such patients are male smokers; some have recent exposure to volatile hydrocarbons, and case reports have additionally identified recent exposure to chlorine and smoked cocaine. This suggests that—when there is susceptibility—inhaled toxic agents enhance pulmonary endothelial damage and thus allow the initiation of autoimmunity or the ready access of existing autoantibody to basement membrane. The usual respiratory presentation is with cough, breathlessness, and haemoptysis, which is intermittent and ranges from occasional streaks to massive fatal bleeding. Systemic symptoms of fever, joint pains, or weight loss occur occasionally but are unusual. The chest radiograph shows patchy or diffuse shadowing due to intra-alveolar blood, usually resolving over the course of 2 weeks unless there is further bleeding. At the time of bleeding there may be arterial hypoxaemia and reduced lung volumes. Serial measurement of TLco can be used to monitor progression. Prolonged bleeding may lead to iron-deficiency anaemia. Renal function may be normal initially and then deteriorate over days to weeks.

Treatment and prognosis

Supportive treatment may be required during acute bleeding, and artificial ventilation is occasionally necessary. Steroids, other immunosuppressant drugs (cyclophosphamide in particular), and plasmapheresis may all be used to control renal disease and are additionally helpful in treating pulmonary haemorrhage. Patients should not smoke (there are case reports of even a single cigarette leading to catastrophic recurrence of haemorrhage), and should avoid hydrocarbon exposure. Prognosis generally depends more on the renal effects than the pulmonary effects. For further discussion see Chapter 21.8.7.

Idiopathic pulmonary haemosiderosis

This is a rare disorder of children and young adults in which there is recurrent alveolar bleeding in the absence of kidney disease. The alveolar blood may provoke a fibrogenic stimulus and the development of diffuse pulmonary fibrosis. Anti-GBM antibody has not been detected, and the electron microscopic appearance of the basement membrane shows no consistent abnormality. Recent studies suggest that many cases may be a consequence of vasculitis at the pulmonary capillary level (pauci-immune pulmonary capillaritis). This is associated inconsistently with P-ANCA activity.

Although termed ‘idiopathic’, the condition is associated with premature birth and an increasing number of environmental exposures. One such that has incited particular interest is to the stachybotrys mould, which may contaminate wet or damp accommodation, and which releases a particularly potent toxin with haemorrhagic properties. This is now thought to cause some childhood cases, perhaps in synergy with environmental tobacco smoke. Associations with rheumatoid arthritis, cow’s milk allergy, and coeliac disease are also recognized (and long-term survivors have an increased incidence of autoimmune disease), but the association with coeliac disease might be a consequence of cow’s milk allergy also rather than gluten intolerance. A number of other environmental causes have been suggested, but the stronger the evidence for their causal roles, the less appropriate is the diagnostic rubric of ‘idiopathic’ pulmonary haemosiderosis, hence these are identified below under the heading ‘other causes’.

Clinical features

Recurrent alveolar haemorrhage is generally manifested by cough with haemoptysis and breathlessness, but haemoptysis is not invariably present, and in children a failure to thrive may be prominent together with the effects of severe chronic iron-deficiency anaemia. Severe acute bleeds are more common in childhood and may be life threatening. Physical examination is unhelpful. The chest radiograph and CT scan show the nonspecific appearances of intra-alveolar blood, which usually clears spontaneously over 1 to 3 weeks. With chronicity the appearances of diffuse pulmonary fibrosis with honeycombing may supervene. Lung function tests then show a progressive loss of volume and reduction of gas transfer, but an unexplained obstructive pulmonary defect occurs occasionally.

Treatment and prognosis

Supportive treatment is the same as for Goodpasture’s syndrome. In addition, there are case reports recording responses to the avoidance of milk and gluten, and to the use of immunosuppressive agents including corticosteroids and cyclophosphamide. Some patients recover spontaneously with or without residual pulmonary damage.

Other causes of diffuse alveolar haemorrhage

Although diffuse alveolar haemorrhage is not a principal or defining feature of disorders other than Goodpasture’s syndrome and idiopathic pulmonary haemosiderosis, it may occur with or complicate a wide variety of disorders with vasculitic, immunological, infective, vascular, haemostatic, toxic, or unknown origins. In many of the cases that have been reported a contributory role could have been played by several different disorders, their complications, and their various treatments.

Vasculitic disorders occasionally cause prominent diffuse alveolar haemorrhage, particularly Wegener’s granulomatosis (granulomatosis with polyangitis) and microscopic polyangiitis. These may simulate Goodpasture’s disease closely, since they commonly cause necrotizing glomerulonephritis. Wegener’s granulomatosis is distinguished from Goodpasture’s syndrome clinically by the common involvement of upper respiratory tract structures, histologically by granulomatous vasculitis, and immunologically by circulating anti-neutrophil cytoplasmic antibodies (ANCA) that are of the cytoplasmic type (C-ANCA) and are directed against proteinase-3 in about 90% of cases. By contrast, microscopic polyangiitis does not typically involve the upper respiratory tract, its vasculitis is not granulomatous, and its anti-neutrophil cytoplasmic antibodies are perinuclear (P-ANCA), directed against the myeloperoxidase of neutrophil cytoplasmic granules. Other vasculitic disorders involving the lung are very rare causes of diffuse alveolar haemorrhage, but include Churg–Strauss syndrome, polyarteritis nodosa, Henoch–Schönlein purpura, and Takayasu arteritis. Diffuse capillary haemorrhage from fulminant vasculitic disease of whatever cause carries a grave prognosis, with mortality of 25 to 50%.

Diffuse alveolar haemorrhage may also arise as an unusual respiratory feature of several nonvasculitic immunological disorders. Most prominent is systemic lupus erythematosus, in which lupus anticoagulant, thrombcytopenia and active nephritis may all play a role, but there are also reports of diffuse alveolar haemorrhage complicating the primary antiphospholipid antibody syndrome, IgA nephropathy, idiopathic membranous nephropathy, scleroderma, renal and bone marrow transplantation, and chronic active hepatitis.

Other reports have implicated lymphangioleiomyomatosis, hymenopteran stings and snake bites, moulds other than stachybotrys and their mycotoxins, infections (AIDS, invasive aspergillosis, dengue fever, cytomegalovirus, leptospirosis, stenotrophomonas, group A streptococci, strongyloidiasis, varicella), familial Mediterranean fever, cardiopulmonary bypass, cold injury, occupational exposure to tri- and pyromellitic anhydride, lymphangiography contrast media, and several medications (amiodarone, azathioprine, carbimazole, nitrofurantoin, mitomycin C, d-penicillamine, rituximab, sirolimus, surfactant therapy, valproate), with many of these conditions involving disseminated intravascular coagulation. The list is completed by causes of chronic pulmonary venous congestion (mitral stenosis, chronic left ventricular failure, pulmonary veno-occlusive disease), certain congenital anomalies of the heart, malignant hypertension, and disorders (or medications) that disrupt bleeding and coagulation mechanisms (thrombocytopenia, leukaemia, thrombinolytic therapy, platelet glycoprotein IIb/IIIa inhibitor, anticoagulant poisoning, clopidogrel, factor V deficiency). Combinations of such factors are commonly found in individual cases, and it may be that important interactions occur, without which the probability of diffuse haemorrhage is remote. Although capillary stress from high pressure gradients is thought to be a major factor underlying diffuse pulmonary haemorrhage in exercising horses (and camels), it appears a rare or unheard cause in most other species, but the use of negative-pressure ventilation in humans has been reported to have a similar effect.

Further reading

Anonymous (2000). From the Centers for Disease Control and Prevention. Update: pulmonary hemorrhage/hemosiderosis among infants—Cleveland, Ohio, 1993–1996. JAMA, 283, 1951–3.Find this resource:

    Bosch X, Guilabert A, Font J (2006). Antineutrophil cytoplasmic antibodies. Lancet, 368, 404–18.Find this resource:

      Collard HR, Schwarz MI (2004). Diffuse alveolar hemorrhage. Clin Chest Med, 25, 583–92.Find this resource:

        Hudson BG, et al. (2003). Alport’s syndrome, Goodpasture’s syndrome, and type IV collagen. N Engl J Med, 348, 2543–56.Find this resource:

          Ioachimescu OC, Sieber S, Kotch A (2004). Idiopathic pulmonary haemosiderosis revisited. Eur Respir J, 24, 162–70.Find this resource:

            Jara LJ, Vera-Lastra O, Calleja MC (2003). Pulmonary-renal vasculitic disorders: differential diagnosis and management. Curr Rheumatol Rep, 5, 107–15.Find this resource:

              Lazor R (2011). Alveolar haemorrhage syndromes. Eur Respir Mon, 54, 15–31.Find this resource:

                Pacheco A, et al. (1991). Long term follow-up of adult idiopathic pulmonary hemosiderosis. Chest, 99, 1525–6.Find this resource:

                  Schwarz MI, Brown KK (2000). Small vessel vasculitis of the lung. Thorax, 55, 502–10.Find this resource:

                    Semple D, et al. (2005). Clinical review: Vasculitis on the intensive care unit-part 1: diagnosis. Crit Care, 9, 92–7.Find this resource:

                      Specks U (2001). Diffuse alveolar hemorrhage syndromes. Curr Opin Rheumatol, 13, 12–17.Find this resource: