Search

Background: In the absence of evidence of arteritis or Wegener's granulomatosis, the syndrome of lung hemorrhage and nephritis has been commonly associated with anti-glomerular basement membrane (GBM) antibodies. However, it has been increasingly recognized that many cases are associated with antineutrophil cytoplasmic antibodies (ANCAs)., Objective: To review available clinical and pathologic findings to determine the diseases accounting for lung hemorrhage and nephritis., Methods: We studied the records of 750 patients from whom serum samples were sent to our laboratory for anti-GBM antibody assays between 1981 and 1993 and found 88 patients with evidence of lung hemorrhage and nephritis. Serum samples were retested, using current methods, for anti-GBM antibodies (against noncollagenous 1 domain of the alpha 3 chain of type IV collagen) and for antibodies to proteinase 3 and myeloperoxidase--the two types of ANCA of diagnostic value., Results: Of 88 patients with evidence of lung hemorrhage and nephritis, 48 had ANCAs, six had anti-GBM antibodies, and seven had both. In 48 patients with ANCAs, the pathologic findings that accounted for the pulmonary renal syndrome were pauci-immune necrotizing and crescentic glomerulonephritis and pulmonary capillaritis. Only eight had convincing evidence (during life) of Wegener's granulomatosis and only one other had documented arteritis. In 27 patients without ANCAs or anti-GBM antibodies, a variety of unrelated renal and pulmonary diseases were found., Conclusions: The largest group of patients who present with the syndrome of lung hemorrhage and nephritis have ANCAs and not anti-GMB antibodies. Appropriate tests for antibodies to proteinase 3, antibodies to myeloperoxidase, and anti-GBM antibodies provide reliable guides for making a diagnosis in patients with this pulmonary renal syndrome.

Three patients with rheumatoid arthritis had bleeding and bruising and were found to have circulating antibodies specifically directed against factor VIII. Review of the literature indicates that this is an unusual but serious complication of rheumatoid arthritis. In two of our three cases, disappearance of the factor VIII antibody coincided with the oral administration of cyclophosphamide; in the third patient, antibody titers declined several weeks after cyclophosphamide therapy was discontinued. Our collected experience involving 11 nonhemophilic patients with factor VIII antibodies indicates that a good response to cyclophosphamide therapy may be expected if the antibody titer is less than 10 Bethesda units at the initiation of treatment. While optimum treatment of these patients is not yet established, we suggest an initial trial of prednisone and, if there is no response, then therapy with cyclophosphamide.

Intra-alveolar hemorrhage is a known complication of lupus erythematosus (LE), but its cause is controversial. Some authors have shown immune complexes (ICs) deposited at various sites in the alveolar septae and postulated that these deposits result in pulmonary hemorrhage (PH). A patient with LE and PH had no detectable IC deposits at a time when IC disease was present in the kidney and vasculitis was active in the skin. Reviewing the literature, we show that IC deposits in the lung are nonspecific and are not correlated with PH. We propose that classification schemes that differentiate between IC-mediated PH and idiopathic PH are arbitrary, and that patients thought to have idiopathic PH should be followed up prospectively to monitor the development of possible immunologic disease.

The response to a highly purified concentrate of porcine factor VIII was evaluated in 45 bleeding episodes in 38 patients with high responding inhibitor antibodies to factor VIII. A total of 437 infusions were given. The patients came from 25 hemophilia centers in the United States. The majority had a life- or limb-threatening hemorrhage for which other modalities had not been successful. In 32 of 45 episodes, a good to excellent response was obtained. Adverse reactions were minimal, occurring in 17 treatment episodes, and were mostly treated with antihistamines and/or hydrocortisone. No clear predictor of clinical response to porcine factor VIII concentrate was identified, including pretreatment human and porcine inhibitor levels, percentage of cross-reactivity between the human and porcine antibodies, and the presence of measurable levels of factor VIII after the porcine factor concentrate was given. Anamnesis to porcine factor VIII did occur in some instances. Porcine factor VIII is a valuable modality in the treatment of serious hemorrhages in patients with inhibitors to factor VIII. Its use should be considered early in the course of severe hemorrhage in these patients.