Microscopic polyangiitis

* Objectives.-To determine the incidence of necrotizing alveolar capillaritis, to elucidate its histopathogenesis and the most reliable histopathologic features for its detection, and to explore its relationship with systemic vasculitis in microscopic polyangiitis.

Methods.-Twenty-five autopsy cases of microscopic polyangtis were examined. Double staining with periodic acid-silver methenamine and trichrome was used.

Results.-Periodic acid-silver methenamine and trichrome staining proved to be a useful and convenient method for the detection of necrotizing alveolar capillaritis. Capillaritis was detected in 10 (40%) of the 25 cases studied. Of 18 cases that exhibited the early degenerative stage of systemic vasculitis, capillaritis was seen in 10 (56%). Fibrinoid necrosis of the alveolar wall was the most reliable histopathologic feature for detecting necrotizing

alveolar capillaritis in microscopic polyangiitis. The early stage of capillaritis displayed very little neutrophil exudation, and conspicuous accumulation of neutrophils developed after fibrinoid necrosis. Statistical analysis revealed that alveolar capillaritis had a statistically significant correlation (P

Conclusion.-Necrotizing alveolar capillaritis was detected in about half of the microscopic polyangiitis autopsy cases that exhibited the early degenerative stage of systemic vasculitis, and results based on earlier reports of clinical observations may have been too conservative. (Arch Pathol Lab Med. 1997;121:144-149)

Polyarteritis nodosa (PAN) is pauci-immune systemic necrotizing vasculitis without granulomatous inflammation, which has usually been subclassified into classic and microscopic forms.l2 Recently, microscopic polyarteritis was alternatively named microscopic polyangiitis (MPA).2 Pulmonary hemorrhage was clinically reported to occur in 20% to 30% of patients with microscopic polyarteritis,3 whereas it has rarely been described in classic

PAN.4

In 1985, Mark and Ramirez5 described five histopathologic criteria for the diagnosis of pulmonary capillaritis in association with lung hemorrhage in patients with systemic vasculitis. Since then, using their criteria, pulmonary capillaritis (also referred to as necrotizing alveolar capillaritis6) has been described in various diseases, including collagen vascular disorders,7-9 although its histopathogenesis has not been elucidated. Necrotizing alveolar capillaritis has been described in biopsy cases of MPA with a clinical diagnosis of pulmonary hemorrhage,lsl7 but, to the best of our knowledge, it has not been reported in cases without clinical features of pulmonary hemorrhage.

In the present study, autopsy cases of MPA were surveyed to determine the incidence of necrotizing alveolar capillaritis and to investigate its relation with systemic necrotizing vasculitis, including the pulmonary vasculature and renal glomeruli. In this report, we reevaluate Mark and Ramirez's five histologic criteria and discuss the histopathogenesis of necrotizing alveolar capillaritis in MPA. For the histologic detection of necrotizing alveolar capillaritis, we used double staining with periodic acidsilver methenamine (PASM) and trichrome, as well as hematoxylin-eosin.

MATERIALS AND METHODS

Thirty-three cases of PAN from our autopsy files were reviewed and evaluated. Two cases of classic PAN, in which necrotizing alveolar capillaritis was not detected, were excluded from this study. Thirty-one cases were classified as MPA, of which six cases were excluded because of bronchopneumonia, fungal infection, or the presence of inclusion bodies suggestive of a pulmonary viral infection, such as cytomegalovirus. The remaining 25 cases had undergone autopsy between 1970 and 1995 at Asahi General Hospital (nine cases), Chiba University Hospital (five cases), Narita Red Cross Hospital (two cases), Chiba Workers' Compensation Hospital (two cases), Kashima Workers' Compensation Hospital (two cases), Teikyo University Ichihara Hospital (one case), Kawasaki Steel Corporation (Chiba) Hospital (one case), Yokaichiba City Hospital (one case), Chiba Social Health Insurance Hospital (one case), and Funabashi Social Health Insurance Hospital (one case). Clinical information was obtained from the records of the attending physicians.

Of these 25 cases, deposition of immunoglobulins (IgA, IgG, and IgM) or complements (Cl and C3) was not demonstrated in lesions of necrotizing vasculitis by indirect immunofluorescence using paraffin sections of postmortem renal or other organ tissues.

Histologic specimens of the 25 cases were examined, and organ distribution, stage, and vessels involved in necrotizing vasculitis were evaluated. No granulomatous lesions were observed in these cases.

Paraffin sections of the lungs were stained with hematoxylineosin and elastic van Gieson's stains. Trichrome stain and phosphotungustic acid-hematoxylin stain were sometimes used to confirm vasculitic changes. Double staining with PASM and trichrome was performed on at least one section with lung hemorrhage in each case. In cases without hemorrhage, sections were chosen randomly.

Immunohistochemical study was performed in the cases with necrotizing alveolar capillaritis using paraffin sections of lung tissues.13 After trypsin digestion, the sections were incubated with rabbit anti-human IgA, IgG, IgM, C1, and C3 antibodies (MBL, Nagoya, Japan), and subsequently reacted with fluorescein isothiocyanate-labeled goat anti-rabbit IgG antibody.

Twenty-five cases were divided into two subgroups, with (subgroup 1, cases 1 through 10) or without (subgroup 2, cases 11 through 25) necrotizing alveolar capillaritis. Statistical analysis was performed using Fisher's exact probability test or the MannWhitney LI test.

RESULTS

Clinical Findings

The 25 MPA cases included 11 men and 14 women. Their ages ranged from 38 to 80 years (average, 64.6 years). None of the patients had a history of bronchial asthma or allergy.

Symptoms suggestive of MPA14 ls included fever in 21 cases; anorexia in 14 cases; general malaise in 12 cases; abdominal pain in 7 cases; weight loss of more than 4 kg since onset in 5 cases; development of mononeuropathy, multiple mononeuropathies, or polyneuropathy in 6 cases; muscle weakness in 4 cases; and myalgia in 2 cases. Respiratory symptoms included cough in 10 cases, sputa in 9 cases, hemoptysis in 2 cases, and dyspnea in 10 cases. Hematuria was noted in 21 cases and proteinuria in 20 cases. More than 1.5 mg/dL of serum creatinine and more than 40 mg/dL of blood urea nitrogen were both noted in 14 cases, indicating renal functional impairment. Fisher's exact probability test revealed no statistical difference (P

Antineutrophil cytoplasmic antibodies (ANCAs) had been examined in cases 5,17, and 23. Perinuclear ANCA (P-ANCA) was positive in cases 5 and 17, whereas cytoplasmic ANCA (C-ANCA) was negative in all three of these cases. The target antigen of P-ANCA in case 17 was determined to be myeloperoxidase by enzyme-linked immunosorbent assay. Reaction for anti-basement membrane antibody in cases 2 and 23 was negative. Serum antibody to hepatitis B surface antigen was detected in case 5, whereas hepatitis B surface antigen was not detected in any case.

Various kinds of antibacterial drugs had been used in 19 cases, but they had had no remarkable effect on the systemic illness. On the other hand, corticosteroids and cyclophosphamide were effective. Although corticosteroids and cyclophosphamide had been used more frequently in subgroup 2 than in subgroup 1, statistical analysis with the Mann-Whitney U test showed no significant difference (P

Hemodialysis or peritoneal dialysis had been performed in case 16 for 4 weeks, in case 21 for 3 weeks, and in cases 2, 11, and 23 for about 2 weeks. Oxygen was administered in cases 2, 4, 5, 7,10,16,17, 23, and 24. An artificial respirator was used for cases 2, 4, 7, 10, and 23. The inspired oxygen concentration was more than 50% in cases 2 (70% for 6 days and 100% for a day), 7 (100% for 3 hours), and 23 (60% for 6 days and 100% for 11 days). These treatments showed no statistical differences between the subgroups (P = .28 for the period of hemodialysis or peritoneal dialysis [Mann-Whitney U test], P = .22 for oxygen administration [Fisher's exact probability test], P = .064 for artificial respiration [Fisher's exact probability test], and P = .32 for the period of inspired oxygen concentration of more than 50% [Mann-Whitney U test]).

Pathologic Findings

Acute pulmonary hemorrhage manifested by numerous intraalveolar red blood cells was seen in all the cases of subgroup 1 and in three cases (11,13, and 14) of subgroup 2. Statistical correlation was seen between pulmonary hemorrhage and necrotizing alveolar capillaritis (Fisher's exact probability test, P = .000087). The intensity of hemorrhage in the cases of subgroup 1 varied, as listed in the Table, and two patients died of pulmonary hemorrhage. The hemorrhage seen in subgroup 2 was focal, and its etiology was uncertain. Hemosiderin-laden cells were seen in the alveolar lumen and interstitium in cases 23 and 24.

Necrotizing alveolar capillaritis was observed in close proximity to acute lung hemorrhage in 10 (40%) of the 25 cases. Fibrinoid necrosis of the alveolar wall was clearly demonstrated by double staining with PASM and trichrome in certain areas of these cases. Fibrinoid material was deposited at the alveolar wall, where small portions of the basal lamina were missing (Fig 1). Fibrin and red blood cells appeared to be spilling into the alveolar lumen via the disrupted segments of necrotic capillaries (Fig 2). In seven cases (Table), fibrin clots were attached to the alveolar septum, which was sometimes discontinuous (Fig 3), in a sessile manner; these often contained erythrocytes. There was very little inflammatory cell exudate, including neutrophils, in these areas. Only rare thrombi were seen in case 6.

In other areas, inflammatory cells, predominantly neutrophils, were seen in and around the necrotic alveolar wall. Prominent accumulation of inflammatory cells, consisting mostly of neutrophils and strands of fibrin, was seen at the necrotic alveolar septal wall (Fig 4, A), in which remnants of the alveolar septum were revealed by PASM-trichrome stain (Fig 4, B). The grade of neutrophil exudation varied widely from case to case and in each lesion of capillaritis, even within the same case (Table).

Mononuclear cells consisting predominantly of histiocytes and lymphocytes were seen in the alveolar interstitium to a mild degree in cases 4 and 5 and to a moderate degree in cases 7 and 10.

Indirect immunofluorescence using paraffin sections of 10 cases with necrotizing alveolar capillaritis showed no deposition of IgA, IgG, IgM, C1, or C3.

Histologic evidence of necrotizing vasculitis was distributed throughout the systemic organs. Necrotizing bronchial angiitis was seen in 58% of the cases, whereas necrotizing pulmonary angiitis was detected in only 16% (four cases in subgroup 1). Necrotizing alveolar capillaritis was statistically correlated with necrotizing pulmonary vasculitis (P = .017), but not with necrotizing bronchial vasculitis. As for the systemic organ distribution of vasculitis, no statistical differences could be determined between cases with and without necrotizing alveolar capillaritis.

The size of arteries involved in systemic necrotizing vasculitis was also evaluated. Interlobar and large arcuate arteries of the kidney and arteries of similar size in other organs were regarded as medium-sized.lb Nine (36%) of 25 MPA cases showed necrotizing vasculitis of mediumsized arteries. There was no statistical correlation between necrotizing alveolar capillaritis and the size of involved blood vessels.

The stage of vasculitis was determined according to Arkin's classification.l7 The early degenerative stage, seen in 10 cases of subgroup 1 and in 8 cases of subgroup 2, had a statistical correlation with necrotizing alveolar capillaritis (P = .013), whereas acute inflammatory, granulation tissue, or healed granulation tissue stages did not. In cases showing the early degenerative stage of systemic vasculitis, alveolar capillaritis was seen in 10 (56%) of 18 cases.

Glomerulitis with necrosis of glomerular tufts and/or crescents was seen in 21 cases. Glomerular changes were divided into the following four categories: glomerular tuft necrosis, cellular, fibrocellular, and fibrous crescent. Statistical analysis revealed no correlation between necrotizing alveolar capillaritis and these glomerular alterations. The P value of Fisher's exact probability test between alveolar capillaritis and glomerular tuft necrosis was .053.

COMMENT

Necrotizing alveolar capillaritis was observed in 10 (40%) of 25 autopsy cases of MPA in the present study. The incidence of alveolar capillaritis was 56% for the cases that exhibited the early degenerative stage of systemic vasculitis. On the other hand, hemoptysis was clinically seen in only two cases. Focal and/or mild degrees of alveolar hemorrhage due to capillaritis seemed to be insidiously occurring irrespective of the clinical symptoms of pulmonary hemorrhage.

In 1985, Mark and Ramirez5 described the following five histopathologic criteria for pulmonary capillaritis: (1) fibrin thrombi occluding capillaries in interalveolar septa; (2) fibrin clots attached to interalveolar septa in a sessile manner; (3) fibrinoid necrosis of capillary walls; (4) neutrophils and nuclear dust in the fibrin, in the interstitium, and in the adjacent alveolar blood; and (5) interstitial red blood cells and interstitial hemosiderin. As has been pointed out previously,18,19 criteria 1 and 5 seem to be less specific. In the present investigation, as shown in the Table, criterion 2 could not be confirmed in three cases. Neutrophils or nuclear dust, as noted in criterion 4, were not always seen at the lesion of necrotizing capillaritis. Therefore, fibrinoid necrosis of alveolar septa seemed to be the most reliable criterion for the detection of necrotizing alveolar capillaritis in cases of MPA.

The histologic determination of necrotizing alveolar capillaritis is difficult owing to the structural characteristics of the lung.5's First, the alveolar wall is so thin that structural remnants could scarcely be identified when involved in a necrotizing process. Second, the alveolar interstitium is so small that most erythrocytes, fibrin, and inflammatory cells possibly would not accumulate at the site of necrotizing capillaritis but, rather, float away into the alveolar airspace. This would decrease the possibility of finding these components at the real site of inflammation.

To deal with these issues, it is important to recognize the distorted alveolar structure. We showed that the structure of alveolar walls was clearly visualized by the use of PASM stain. Discontinuity of the alveolar wall could be clearly distinguished, and the possibility of finding remnants of the alveolar wall also seemed to be greater. Moreover, deposition of fibrin or fibrinoid material was demonstrated with trichrome stain, usually in connection with disruption of the alveolar wall. This double staining, a method often used for renal biopsy specimens, proved to be a useful and convenient method for the recognition of necrotizing alveolar capillaritis.

As for the histologic features of necrotizing alveolar capillaritis seen in the present cases, fibrinoid material was usually deposited at the disrupted alveolar wall, although the amount varied. Exudation of neutrophils at the site of necrotizing alveolar capillaritis also varied widely among the cases and even within the same case. Because neutrophils, as well as other inflammatory components, might spill away from the disrupted capillary into alveolar lumina, it was difficult to determine the exact pathologic process of necrotizing alveolar capillaritis. However, in the cases we studied, fibrinoid necrosis with very little exudation of inflammatory cells seemed to precede the prominent accumulation of neutrophils, which could be in accordance with the systemic vasculitis of MPA.20

The incidence of pulmonary arterial involvement in MPA or microscopic polyarteritis is obscure, although it has been reported that pulmonary vasculitis occurs only occasionally (in less than 10%) in PAN cases that included both its classic and microscopic forms.zl In our cases of MPA, pulmonary arterial involvement was seen in 16%, and statistical analysis revealed a correlation with necrotizing alveolar capillaritis. This correlation seemed to be due to the anatomical fact that alveolar capillaries are an internal part of the pulmonary circulatory system.

Statistical analysis also demonstrated a correlation between necrotizing alveolar capillaritis and the early degenerative stage of MPA, suggesting that alveolar capillaritis occurred when systemic vasculitis was exhibiting early degenerative changes.

Antineutrophil cytoplasmic antibodies have been detected in various diseases.2223 In association with MPA, P-ANCA has been reported more often than C-ANCA.24 In the present cases, P-ANCA was positive in two of the three cases examined for these antibodies. Although ANCAs were not examined in our older cases, it is important to investigate ANCAs and their target antigens in patients with systemic vasculitides, especially in those presenting with pulmonary hemorrhage,25 because it has been suggested that these autoantibodies may be involved in the pathogenesis of necrotizing alveolar capillaritis as well as systemic vasculitis.26

The authors are grateful to the following pathologists for allowing us to study their cases: Yoshio Suzuki, MD, Asahi General Hospital, Chiba; Osamu Matsuzaki, MD, Chiba Cancer Center; Kazuhiko Yasumi, MD, Narita Red Cross Hospital, Chiba; Akio Konno, MD, Chiba Workers' Compensation Hospital; Hirosato Iwase, MD, Kashima Workers' Compensation Hospital, Ibaraki; Isamu Sugano, MD, and Koichi Nagao, MD, Teikyo University Ichihara Hospital, Chiba; Hiroshi Horie, MD, Chiba Children's Hospital; and Haruo Ohkubo, MD, Funabashi Social Health Insurance Hospital, Chiba.

References

1. Fauci AS, Haynes BF, Katz P. The spectrum of vasculitis: clinical, pathologic, immunologic, and therapeutic considerations. Ann Intern Med. 1978;89:660676.

2. Jennette JC, Falk RJ, Andrassy K, et al. Nomenclature of systemic vasculitides: proposal of an international consensus conference. Arthritis Rheum. 1994;37:187-192.

3. Haworth SJ, Savage COS, Carr D, Hughes JMB, Rees AJ. Pulmonary haemorrhage complicating Wegener's granulomatosis and microscopic polyarteritis. Br Med 1985;290:1775-1778.

4. Leatherman JW, Davies SF, Hoidal JR. Alveolar hemorrhage syndromes: diffuse microvascular lung hemorrhage in immune and idiopathic disorders. Medicine.1984;63:343-361.

5. Mark EJ, Ramirez JF. Pulmonary capillaritis and hemorrhage in patients with systemic vasculitis. Arch Pathol Lab Med.1985;109:413-418. 6. Bosch X, Font J, Mirapeix E, Revert L, Ingelmo M, Urbano-Marquex A. Antimyeloperoxidase autoantibody-associated necrotizing alveolar capillaritis. Am Rev Respir Dis.1992;146:1326-1329.

7. Travis WD, Colby TV, Lombard C, Carpenter HA. A clinicopathologic study of 34 cases of diffuse pulmonary hemorrhage with lung biopsy confirmation. Am J Surg Pathol. 1990;14:1112-1125.

8. Travis WD, Koss MN. Vasculitis. In: Dail DH, Hammar SP, eds. Pulmonary Pathology. 2nd ed. New York, NY: Springer-Verlag; 1994:1027-1095. 9. Miller RR. Diffuse pulmonary hemorrhage. In: Thurlbeck WM, Churg AM, eds. Pathology of the Lung. 2nd ed. New York, NY: Thieme Medical Publishers; 1995:365-373.

10. Schwarz Ml, Mortenson RL, Colby TV, et al. Pulmonary capillaritis: the association with progressive irreversible airflow limitation and hyperinflation. Am Rev Respir Dis. 1993;148:507-511.

11. Gal AA, Salinas FF, Staton GW. The clinical and pathological spectrum of antineutrophil cytoplasmic autoantibody-related pulmonary disease: a comparison between perinuclear and cytoplasmic antineutrophil cytoplasmic autoantibodies. Arch Pathol Lab Med. 1994;118:1209-1214.

12. Imoto EM, Lombard CM, Sachs DPL. Pulmonary capillaritis and hemorrhage: a clue to the diagnosis of systemic necrotizing vasculitis. Chest. 1989;96: 927-928.

13. Choi YJ, Reiner L. Immunofluorescence of renal lesions in paraffin-embedded and fresh-frozen sections. Am I Clin Pathol. 1980;73:116-119. 14. Savage COS, Winearls CG, Evans DJ, Rees AJ, Lockwood CM. Microscopic polyarteritis: presentation, pathology and prognosis. QJ Med.1985;56:467-483.

15. Lightfoot RW Jr, Michel BA, Bloch DA, et al. The American College of Rheumatology 1990 criteria for the classification of polyarteritis nodosa. Arthritis Rheum. 1990;33:1088-1093.

16. Heptinstall RH. Polyarteritis (periarteritis) nodosa, Wegener's syndrome and other forms of vasculitis. in: Heptinstall RH, ed. Pathology of the Kidney. 4th ed. Boston, Mass: Little, Brown & Co; 1992:1097-1162.

17. Arkin A. A clinical and pathological study of periarteritis nodosa: a report of five cases, one histologically healed. Am J Pathol.1930;6:401-431.

18. Sieber SC, Cuello B, Gelfman NA, Garfinkel HB. Pulmonary capillaritis and glomerulonephritis in an antineutrophil cytoplasmic antibody-positive patient with prior granulomatous aortitis. Arch Pathol Lab Med. 1990;114:1223-1226. 19. Myers JL, Colby TV, Yousem SA. Common pathways and patterns of injury. In: Dail DH, Hammar SP, eds. Pulmonary Pathology. 2nd ed. New York, NY: Springer-Verlag; 1994:57-77.

20. Rosen S, Falk RJ, Jennette JC. Polyarteritis nodosa, including microscopic form and renal vasculitis. In: Churg A, Churg J, eds. Systemic Vasculitides. New York, NY: Igaku-Shoin Medical Publishers; 1991:57-77.

21. Lie JT. Systemic and isolated vasculitis: a rational approach to classification and pathologic diagnosis. Pathol Annu. 1989;24:25-114.

22. Jennette JC, Falk RJ. Disease associations and pathogenic role of antineutrophil cytoplasm autoantibodies in vasculitis. Curr Opin Rheumatol. 1992;4:915.

23. Kallenberg CGM, Brouwer E, Weening JJ, Cohen Tervaert JW. Anti-neutrophil cytoplasmic antibodies: current diagnostic and pathophysiological potential. Kidney Int. 1994;46:1-15.

24. Gross WL, Schmitt WH, Csernok E. ANCA and associated diseases: immunodiagnostic and pathogenetic aspects. Clin Exp Immunol. 1993;91:1-12.

25. Gaudin PB, Askin FB, Falk RJ, Jennette JC. The pathologic spectrum of pulmonary lesions in patients with anti-neutrophil cytoplasmic autoantibodies specific for anti-proteinase 3 and anti-myeloperoxidase. Am J Clin Pathol. 1995;1 04:7-16.

26. Beer DJ. ANCAs aweigh. Am Rev Respir Dis. 1992;146:1128-1130.

Copyright College of American Pathologists Feb 1997
Provided by ProQuest Information and Learning Company. All rights Reserved