Correlation of antineutrophil cytoplasmic antibodies with the extrarenal histopathology of Wegener's (pathergic) granulomatosis and related forms of vasculitis.

We studied the histologic findings from extrarenal biopsies (especially of the lung or upper respiratory tract) or autopsies of 68 patients who were tested for serum antineutrophil cytoplasmic antibodies (ANCAs). We used antigen-specific assays to detect antibodies against proteinase 3 (PR3) and myeloperoxidase (MPO), the two types of ANCAs of proven diagnostic value for the spectrum of diseases that includes Wegener's (pathergic) granulomatosis, microscopic polyarteritis (microscopic polyangiitis), Churg-Strauss syndrome, idiopathic necrotizing and crescentic glomerulonephritis, and their variants. Twenty-eight patients had antibodies to PR3 and 16 had antibodies to MPO; no patient had antibodies to both. All 44 patients with ANCAs had histologic evidence of this spectrum of diseases. Thirteen patients without histologic evidence of this spectrum of diseases had negative tests for ANCAs. There were no pathologic features that reliably identified patients with one or the other type of ANCA. Eighteen of 31 patients with lesions of Wegener's granulomatosis had antibodies to PR3, seven had antibodies to MPO, and six had neither. Three of four patients with necrotizing arteries without granulomas had anti-MPO antibodies, but similar lesions were seen, together with extravascular granulomas, in three patients with anti-PR3 antibodies. Of 16 patients with alveolar hemorrhage, nine had anti-PR3 and five had anti-MPO antibodies. Two patients diagnosed clinically as having Churg-Strauss syndrome had anti-MPO antibodies. In 16 of the 25 patients with ANCAs and a histologic diagnosis of Wegener's granulomatosis the diagnosis was made on the basis of extravascular granulomatous lesions alone, which argues against the requirement for vasculitis. Of six patients with negative tests for ANCAs and histologically diagnosed Wegener's granulomatosis, none had evidence of renal involvement. We conclude that in the appropriate clinical setting the presence of anti-PR3 or anti-MPO antibodies provides reliable evidence of the above spectrum of diseases, but that subclassification (to the extent this is possible) depends on the presence of distinctive clinical or pathologic features. In patients with negative tests for ANCAs, interpretation of clinical and histologic findings remains the only definitive method of diagnosis.