Jump to: Page Content, Site Navigation, Site Search,
You are seeing this message because your web browser does not support basic web standards. Find out more about why this message is appearing and what you can do to make your experience on this site better.
BMJ 2006;333:692-693 (30 September), doi:10.1136/bmj.38978.380000.68
Paul Emery, head1, Jane Freeston, specialist registrar in rheumatology1
1 Academic Section of Musculoskeletal Disease, Chapel Allerton Hospital, Leeds LS7 4SA
Correspondence to: P Emery p.emery{at}leeds.ac.uk
Adult onset Still's disease is a difficult clinical diagnosis. There are few, if any, diagnostic features, with the possible exception of a very high ferritin concentration. Sivakumar and colleagues describe a not atypical case history of a patient presenting with fever.1 Not unnaturally, the patient was assumed to have an infection. This is understandable and correct, as missing infection could clearly have had serious consequences. For every patient with adult onset Still's disease there are a significant number of patients with severe infection, and it is crucial that infection is the first diagnosis to be excluded.
The persistence of the fever after treatment was the first clue that the patient may have something other than infection. Development of a rash was perhaps the first concrete clue to her diagnosis. Her raised ferritin concentration was highly suggestive, although very high concentrations are more specific. The problem with the Yamaguchi diagnostic criteria for diagnosis is that they are quite non-specific. The presence of spiking fever and rash and the deterioration in cardiac function should perhaps have triggered a trial of immunosuppression with steroids. Her doctors were correct to meet the request for a second opinion; these cases require as much expertise and experience as is available.
|
Serum ferritin concentrations above a threshold of 1000 µg/l (five times the upper limit of normal) have been used to suggest adult onset Still's disease in most studies.2 However, similar concentrations can be produced by infection, liver disease, haemophagocytic syndrome, and malignancy. A French study of the validity of hyperferritinaemia as a diagnostic tool reported that a ferritin concentration five times the upper limit of normal had a specificity of 41% and a sensitivity of 80%.3 Others have suggested that concentrations above 3000 µg/l are diagnostic, but only after infection has been excluded.4
Another approach has been to look at the fraction of glycated ferritin, which in healthy subjects constitutes 50-80% of ferritin. In inflammatory diseases, however, saturation of glycosylation mechanisms causes the glycated fraction to drop to 20-50%, and in Still's disease it is often less than 20%. Fautrel and colleagues showed that when this diagnostic criterion was applied in conjunction with a fivefold rise in ferritin, the specificity of this combination rose to 93% but with a fall in sensitivity to 43%.3
The mechanism by which ferritin is raised is probably related to the cytokines released in Still's disease, and therapeutic studies are investigating the role of blocking interleukin 6 and 18 as well as anakinra, the interleukin 1 receptor antagonist.5 These should allow much more effective targeted therapy. New drugs are urgently needed because many patients with this condition respond poorly to low dose methotrexate, the standard drug for arthritis, and high doses produce a high incidence of side effects, including serious infection.
What can you learn from this BMJ paper? Read Leanne Tite's Paper+