Temporal arteritis: cell composition and the possible pathogenetic role of cell-mediated immunity

[Physiopathological data of Horton's syndrome]

GREAT PROGRESS: The understanding of the physiopathology of Horton's syndrome has been made in view of the prevalence of the disease, the access to affected tissue and new molecular biology techniques. And it is now possible to specify the intricacy between the genetic, immunological and vascular components. HORTON'S SYNDROME, A GENETIC DISEASE: The preferential association of the disease with some alleles pf the HLA DR4 group has helped to emphasize the fundamental role of a few amino acids of the second hypervariable area of the HLA-DR molecule. An immunogenetic predisposition appears favourable, or even necessary, for the development of the disease. HORTON'S SYNDROME, AN IMMUNOLOGICAL DISEASE: The temporal arteries of patients presenting with Horton's syndrome are infiltrated by polymorphous inflammatory cells, fundamentally including CD4 T-cell lymphocytes, macrophages, and a few giant cells. Some CD4 lymphocytes have undergone clonal proliferation and show signs of recent antigenic recognition. A fraction of them secrete interferon gamma, which plays a crucial role in the onset, maintenance and orientation of the immunological response. The macrophages play multiple roles notably in maintaining the inflammatory reaction, but also in the destruction of certain structures of the arterial wall. HORTON'S SYNDROME, A VASCULAR DISEASE: The destruction of the arterial wall at the acute stage of the disease appears responsible, sometimes later on, for aneurismal dilatations observed on the aorta of certain patients. Moreover, intimal hyperplasia (mediated by the PDGF (platelet derived growth factor) A and B) and thrombosis (related to the inflammation) join up in reducing the arterial flow and enhancing the risk of ischemic complications during the acute stage. PATHOGENESIS, HYPOTHESES: Various candidate-antigens have been incriminated in the onset of the inflammatory reaction during Horton's syndrome. Some epidemiological and molecular studies are in favour of an exogenous antigen, possibly infectious, but no evidence has been demonstrated in the studies published. A parietal, endogenous antigen might also be at the origin of the cascade of events described during this disease.

Vessel wall morphometry in giant cell arteritis

OBJECTIVE

To investigate morphometrically the relationship between the degree of inflammatory reaction and arterial cross-sectional dimensions in giant cell arteritis (GCA).

METHODS

The media and intima of cross-sectioned inflamed arteries from GCA patients were outlined. The media circumference and the media and intima cross-sectional areas were measured. The two segments in every biopsy displaying the least and most widespread media inflammation were compared, using a paired Student's t-test. Moreover, paired comparisons were made of the intima area in inflamed and noninflamed sectors in single cross-sections.

RESULTS

The segment in each biopsy showing the most widespread media inflammation had the largest circumference and volume of media and intima; the intima cross-sectional area increased disproportionately. The intima was thickest in inflamed sectors in single cross-sections.

CONCLUSION

The close spatial correlation between inflammatory media infiltration and the marked intimal expansion supports the contention that promoting factors produced by inflammatory cells play a pathogenetic role in GCA.

Current concepts in giant cell (temporal) arteritis

Giant cell (temporal) arteritis continues to be a sight-threatening, systemic vasculitis with a poorly understood pathogenesis. The characteristic granulomatous inflammation of the vessel wall commonly leads to local ischemia. Recent advances in immunological investigations have characterized the cellular components of the disease process, but the etiology has so far remained unresolved. A reappraisal of the clinical features of giant cell (temporal) arteritis demonstrates the heterogeneity of the manifestations of the disease, including ischemic optic neuropathy. A range of new laboratory investigations and blood flow studies with color Doppler imaging have demonstrated promising roles, with respect to diagnosis and long-term follow-up. Prompt diagnosis and expeditious treatment require a high index of clinical suspicion, particularly for atypical cases. Corticosteroids remain the treatment of choice, other immuno-suppressive agents being used as second line steroid-sparing agents. Giant cell (temporal) arteritis leads to increased vascular and visual morbidity and, if untreated, may prove fatal. To maintain high standards of management of this enigmatic disorder, ophthalmologists need to be aware of the clinical spectrum of giant cell (temporal) arteritis and currently available diagnostic tests and treatment strategies.

Giant cell arteritis and polymyalgia rheumatica

Giant cell arteritis and polymyalgia rheumatica are linked conditions that frequently occur in the same patient. They are more common in northern Europe and persons of European descent than in other populations. Recent investigations have begun to provide information about the pathogenesis of both syndromes. Both respond to corticosteroids but at different dose levels. Although a number of vascular complications may occur, the outlook is excellent.

[Immunohistochemical study of lesions in Horton's temporal arteritis before and during corticotherapy]

The immunohistochemical study was performed on temporal artery biopsies from eight patients with giant cell (temporal) arteritis: three before treatment, four after a short period of corticosteroid therapy (from 1 day to 7 days) and one during relapse occurring after a treatment of 9 years; from four subjects with clinical symptoms but without histological features of giant cell arteritis and from five negative controls. Before treatment, biopsies of patients with temporal arteritis showed an inflammatory infiltrate with macrophages and T cells, essentially CD4+ and memory T cells (CD45 RO+), expressing the markers of activation IL2 receptor and HLA DR. Few B and NK cells were also detected. Adhesion molecules, LFA1 and I-CAM1, were strongly expressed by T cells and macrophages. In contrast, the ligand to the CD2, the CD58 marker, was rarely detected. These immunohistochemical features were also observed after a short corticosteroid treatment (by intravenous methylprednisolone or oral prednisone), with presence of activated T cells, memory T cells, macrophages and I-CAM1 and LFA1 expressing cells in the infiltrate. A temporal biopsy, performed after a long time of corticosteroid therapy, showed activated T cells, macrophages and memory cells in o,ne arteriole. In controls, this study showed some mononuclear cells dispersed in intima and adventia, but without activated or memory T cells. Our results support the presence of immune local response in temporal arteritis, incompletely improved by a short corticosteroid treatment.

Leukocyte infiltration in synovial tissue from the shoulder of patients with polymyalgia rheumatica. Quantitative analysis and influence of corticosteroid treatment

OBJECTIVE

To investigate the immunologic features of synovitis in patients with polymyalgia rheumatica (PMR) and to assess the modifications induced by corticosteroids.

METHODS

Arthroscopic biopsies of shoulder synovium were obtained from 12 patients with untreated PMR and from 7 patients with PMR that had been treated. Immunohistochemistry was performed on frozen sections utilizing a panel of monoclonal antibodies and computerized image analysis.

RESULTS

Synovitis was present in 10 of 12 (83%) untreated patients and in only 2 of 7 (29%) treated patients. The synovitis was characterized by vascular proliferation and leukocyte infiltration. Infiltrating cells consisted predominantly of macrophages and T Lymphocytes. Almost all T lymphocytes were CD45RO positive. A few neutrophils, but no B cells, natural killer cells, or gamma/delta T cells were found. Intense expression of HLA class II antigens (DR moreso than DP moreso than DQ) was found in the lining layer cells as well as in macrophages and lymphocytes. DR, but not DP or DQ, was expressed by the endothelium of a few vessels. Class II antigen expression correlated with the number of macrophages and lymphocytes. Macrophage infiltration of arteriole walls was observed in 1 untreated patient without giant cell arteritis (GCA). In untreated patients, there was a positive correlation between the percentage of infiltrating T cells and the duration of disease. Steroid therapy was associated with a significant reduction in the number of blood vessels and of HLA class II expression. One treated patient who no longer had symptoms of PMR still had active synovitis: a relapse occurred 4 months after the biopsy.

CONCLUSION

Our findings support the hypothesis that synovitis is a major cause of the musculoskeletal symptoms of PMR. There are immunologic similarities with the vascular inflammation observed in GCA. Corticosteroids act on both the vascular and cellular components of synovitis.

Production of interleukin 6 by granulomas of giant cell arteritis

IL-6 serum levels are increased in patients suffering from GCA, and this cytokine may play a role in the systemic symptoms associated with this disease. We analyzed by in situ hybridization and immunohistochemistry whether IL-6 production originates from arterial granulomas in GCA. Seven arterial biopsy specimens taken from patients with histologically proved GCA were studied. IL-6 production was detected in all cases at both the mRNA and the protein levels. IL-6-producing cells were distributed in the intima, in the media, and in the adventitia. Positive cells were clearly enriched in the media, however, and particularly in contact with the internal elastic lamina. The morphology of the IL-6-producing cells showed that they belonged to several cell populations. In the media, most IL-6-producing cells were macrophages, whereas fibroblasts also produced IL-6 in the intima. Neither endothelial cells nor giant cells were found to express the IL-6 gene. Increased IL-6 serum concentrations returned to normal levels in most patients after administration of corticosteroids, indicating that inhibition of IL-6 production by GCA granuloma cells may be one of the mechanisms of action of corticosteroids in this condition. Production of IL-6 in abnormal arteries may thus participate to the systemic manifestations of GCA.

Headache to worry about

The headache to worry about is one that is unique or different from headaches which the patient has suffered in the past. The association with the headache of meningismus or of focal neurologic symptoms of oculoparesis, other cranial nerve palsies, hemiparesis, or loss of consciousness are particularly worrisome, especially if onset is recent and acute. Headaches related to arteritis or vasculitis usually have a slower subacute course but may also produce focal neurologic deficits. For subjects over 50 years old, temporal arteritis is always a consideration and any new type of headache requires testing of the sedimentation rate to rule out this treatable but potentially devastating problem.

Ultrastructural analysis of asteroid bodies: evidence for membrane lipid bilayer nature of components

Electron microscopic examination of multiple asteroid bodies (ABs) from pulmonary tissue revealed that the core and spokes of these structures are composed of phospholipid bilayers organized in lamellar and tubular forms. These morphologic patterns followed the ones described previously for the lipid-water bulk interface systems. On light microscopy the ABs appeared surrounded by a halo of vacuoles corresponding ultrastructurally to loosely arranged myelin membranes. Transitional forms, between the tightly arranged phospholipid membranes forming the spokes and the loose myelin membranes surrounding them, were seen. Myelin membranes were seen in the cytoplasm of multinucleated giant cells, independent of ABs, in close association with intracellular spaces (lumina) and lined by abundant microvilli identical to the surface microvilli. These spaces were present in central areas of the cell and probably represent surface membrane remnants of partially fused cells. No centrioles, true microtubules, or collagen were seen in the multiple ABs examined. Immunoperoxidase stains for vimentin and tubulin were also negative in ABs. No calcium or phosphorus was seen by X-ray microanalysis. Thus the morphologic evidence is supportive of the hypothesis that, during or after the process of fusion of activated macrophages, the resulting excess of membranes (surface and lysosomal) contributes to the formation of ABs. No evidence supporting the notion that ABs are formed by the aggregation of preexisting cytoskeletal components was found.

Immunological studies in giant cell arteritis

The results of investigations on the humoral immunological mechanisms are conflicting in giant cell arteritis (GCA) and have not been able to explain the pathological findings in the inflamed arterial wall. Altogether, immunological studies suggest that a cell-mediated immune reaction, possibly against an autologous antigen, occurs locally in the arteritic lesions of GCA. The excellent effect of treatment with glucocorticosteroids on the inflammation in GCA can also be explained by this model. The glucocorticosteroids inhibit the synthesis of interleukin-1 (IL-1) by the macrophages and suppress the IL-2 production from the T cells (Palacios, 1982). The observed HLA-DR expression in the arterial wall can be accounted for by the sum of macrophages and activated T cells, the macrophages being the most probable antigen-presenting cells. The interdigitating reticulum cells observed in some of the GCA patients may also be involved in antigen presentation. What the antigen(s) may be is, however, still unknown, as are the factors initiating the inflammatory process. It has recently been possible to extract T lymphocytes from the inflamed tissue and to culture these cells in vitro. After culture, it is possible to study the gene for the T-cell receptor, and probably even the antigenic specificity of the T cells. I hope that this approach may lead to a better understanding of the pathogenic mechanisms in GCA.

Surgical pathology of cranial arteritis and polymyalgia rheumatica

In both clinical and histological terms cranial arteritis is one of the most distinctive of all vascular disorders. The dense granulomatous inflammatory infiltrates which characterize the acute stages of the disease resemble those of Takayasu's arteritis or granulomatous angiitis of the central nervous system, but the clinicopathological features in patients with positive temporal artery biopsies are diagnostic. Well over a third of patients with classical signs and symptoms of cranial arteritis have negative temporal artery biopsies, and focal involvement of arteries of the head and neck is the probable explanation for this. Pathologists should be aware of the wide spectrum of histological changes that occur in muscular arteries as part of normal ageing and must not interpret these as evidence of healed arteritis. The histological changes of healed arteritis include medial chronic inflammation with ingrowth of new blood vessels, focal medial scarring and a bizarre pattern of intimal fibrosis. Although ultrastructural and immunohistochemical studies have provided some insight into the underlying pathological changes, they have not contributed directly to the diagnosis of cranial arteritis. Between 15 and 55% of patients with polymyalgia rheumatica have positive temporal artery biopsies, but apart from an elevated ESR there are no other laboratory investigations or biopsy procedures that contribute to diagnosis.

Temporal artery morphology and morphometry in giant cell arteritis

Paraffin sections and more than 7,000 plastic cross-sections of temporal arteries from 27 patients with a clinical diagnosis of polymyalgia rheumatica (PMR), 16 patients with a clinical diagnosis of temporal arteritis (TA) and 18 age- and sex-matched postmortem controls were studied using light microscopy. A new method was developed to permit a morphometric comparison between biopsies and autopsy specimens. Two stages of inflammation were discerned in TA. In atrophic arterial segments there was a focal, foreign-body, giant-cell reaction to the calcified internal elastic membrane (IEM) with a spatially correlated, mononuclear cell infiltration. Isolated giant cells were also found to attack the IEM in these cases. The majority of the biopsies displayed a different picture with a diffuse macrophage attack on media and intima with numerous and apparently macrophage-derived giant cells, which did not attack calcifications. The latter arteries were significantly widened (p less than 0.02), which indicates that this phase is the later one. Non-inflamed segments of PMR vessels displayed a significant, non-reactive media atrophy compared with controls (p less than 0.03) and their IEM calcifications were significantly larger (p less than 0.02). The circumference was smaller in atrophic PMR arteries than in inflamed TA arteries (p less than 0.006), which contradicts post-inflammatory scarring.

Analysis of adhesion molecules in the immunopathogenesis of giant cell arteritis

To explore the role of adhesion molecules in mediating mononuclear cell localisation, development of the granulomatous reaction, and cell mediated damage to the arterial wall in giant cell arteritis, 17 temporal artery biopsy specimens were examined. Eleven showed the histological features of giant cell arteritis and six showed no evidence of arteritis. All were examined for the expression of LFA-3, ICAM-1 and its receptor LFA-1, and HLA-DR. Temporal arteries with early features of arteritis, as well as histologically unaffected skip areas, showed a regional induction of ICAM-1 expression, but not HLA-DR, on smooth muscle cells of the media. ICAM-1 expression was detected in areas where a clinically important mononuclear cell infiltrate had not yet developed. In more florid cases of giant cell arteritis there was an additional widespread induction of ICAM-1 expression on intimal myofibroblasts. Strong expression of ICAM-1, HLA-DR, and LFA-3 was found on macrophages, epithelioid cells, and giant cells comprising the granulomatous lesion. The pattern of expression of these adhesion molecules suggests that they have a role in leucocyte traffic into the vascular lesion as well as in mediating the intercellular interactions which constitute the granulomatous response.