Effect of iron dextran, gold thiosulphate, and hydrocortisone acetate on experimental synovitis in the guinea pig.

Considerable controversy has revolved about the nature of the disturbed iron metabolism in patients with rheumatoid arthritis and the question whether the depositions of iron in the synovial tissues of these patients represent beneficial or harmful events (Mowat and Hothersall, 1968). Lawson, Owen, and Mowat (1967) showed that there was an increase in the urinary excretion of iron in patients with rheumatoid arthritis compared with controls after the administration of the chelating agent desferrioxamine B. It was considered that these results provided evidence of increased iron storage in these patients. Subsequently Mowat and Hothersall (1968) and Senator and Muirden (1968) showed considerably higher concentrations of iron in the synovial tissue of patients with rheumatoid arthritis compared with controls. Since there is proliferation of the synovial tissue in these patients, relatively large quantities of iron could be deposited in this tissue. The iron is present in the synovial tissue in the form of ferritin and has a distinctive appearance under light or electron microscopy (Muirden, 1966). A similar increase in the quantity of ferritin and in its distribution to that found in patients with rheumatoid arthritis has been noted in experimental arthritis induced in rabbits with caragheenin, Freund's adjuvant, and tuberculin (Muirden and Peace, 1969). Muirden (1966, 1970) and Muirden, Fraser, and Clarris (1967) suggested that the ferritin was deposited in the synovial tissue as the end-result of repeated small haemorrhages and subsequent erythrophagocytosis by the synovial lining cells. This mechanism could possibly explain the findings in rheumatoid arthritis as red blood cells in small numbers can be found in many joint effusions in these patients. However, it is less easily invoked as the mechanism involved in the experimental arthritides which are of short duration. A reduced clearance of Fe59-labelled erthrocytes from the knee joints of rabbits with caragheenin arthritis compared with controls (Muirden, 1969) with subsequent deposition of iron in the synovial tissue has been described (Muirden, Peace, and Rogers, 1969). This confirms the work of Roy and Ghadially (1966), who demonstrated the ability of rabbit synovial tissue to form ferritin from erythrocytes, and of Muirden and others (1967), who showed the formation of ferritin by human synovial cells in culture on exposure to haemoglobin. However, it is possible that the iron might reach the synovial tissue by transfer from the plasma or synovial fluid. The presence of a low serum iron level and an equally low synovial fluid iron (Senator and Muirden, 1966; personal observations) does not exclude the possibility of an iron gradient. Rabbit synovial tissue has also been shown to form ferritin from iron-dextran (Ball, Chapman, and Muirden, 1964). Thus it is clear that synovial tissue has the capacity to form ferritin from haemoglobin or elemental iron, but the source of the iron has not been determined. Muirden (1966) and Muirden and others (1969) have suggested that the presence of iron in the synovial tissue might be harmful and might lead to the release of lysosomal enzymes, resulting in damage to the joint. There is no evidence for this mechanism. In addition, iron stored in the synovial tissue in haemochromatosis is not the cause of the associated arthropathy, which is essentially degenerative in nature (Hamilton, Williams, Barlow, and Smith, 1968; Wardle and Patton, 1969). Richmond, Roy, Gardner, Alexander, and Duthie (1958) showed that a 5 g. course of saccharated oxide of iron, given intravenously to patients with rheumatoid arthritis, was followed not only by an increase in the haemoglobin concentration but also by a