Send letter to journal:
Re: The definition of disease duration in ankylosing spondylitis: comment on the article by Davis et al.

Dear Editor,

Davis et al [1] emphasize the necessity of a uniform definition of the term "disease duration" for the case of ankylosing spondylitis (AS) because different definitions have been used in the past. Besides the duration since disease onset (time of first symptoms), the duration since the time of diagnosis of AS has also sometimes been named "disease duration" [2]. We very much support the initiative for a uniform definition which would indeed ensure comparability across studies and thus facilitate future research.

The conclusion in [1] that "important components of the recommendations to better define duration of disease were found to be :

(1) time since onset of axial AS-specific symptoms (inflammatory back pain),

(2) onset of signs of spondyloarthritis, peripheral or extra-articular symptoms,

(3) onset of associated SpAs, and

(4) time since diagnosis of AS by a health-care provider" does, however, not contribute to a uniform definition.

In two recent surveys [3–5] we found that up to 48 years may lie between first symptoms of AS and the diagnosis of AS, with an average delay in diagnosis of 8.9 years and 8.8 years, respectively, in these surveys.

As an AS researcher and a rheumatologist, both with AS ourselves, we want to emphasize that using the time since diagnosis as "disease duration" does not only disregard many years of the patients' suffering from an undiagnosed but nevertheless very interfering disease. More important: This definition may lead to severe scientific errors in epidemiologic studies. If, for instance, the dependence of the extent of ankylosis on the disease duration is investigated (as in [4, 6]) and the duration between first complaints and diagnosis were neglected, an extended ankylosis found already at "disease duration" = 0 would not at all be surprising. The growth of syndesmophytes does not wait until the disease is diagnosed. If, as another example, the annual incidence of vertebral fractures and its dependence on the disease duration is investigated as in [7], a "disease duration" neglecting the disease duration before the diagnosis of AS, may likewise lead to wrong conclusions, and vertebral fractures which are in fact a result of many years with the disease may be found at negative "disease durations". These examples may illustrate why we regard a “disease duration” of AS neglecting the years before its diagnosis as being unscientific, in contrast to the authors of [1] who mention that ”the time since diagnosis of AS will aid in issues related to regulation, research, and education”.

The authors of [1] also argue that assessing disease duration by studying patient histories from medical records may lack accuracy because of missed or inappropriate diagnosis, and that this might be true for patient's recall of symptoms as well. However, replacing a possibly inaccurate value by a value which is definitely wrong by many years, makes it even worse.

We agree that the time of diagnosis should be collected in addition – besides the time of first symptoms – when data for research is collected. We regret that this is not done, for instance, in the national database of the German collaborative arthritis centres [8] where an evaluation of the delay in diagnosis of AS is therefore impossible. It is also not possible by use of this database to verify the interesting dependence of the female percentage among AS patients on the decade in which the diagnosis was made [4].

A quantity like the disease duration cannot be defined by four alternative and contradictive definitions. We would expect a clear-cut definition like, for instance: (1) In general, "disease duration" of AS is defined as the duration since onset of AS-specific symptoms (inflammatory back pain). (2) If onset of peripheral signs of spondyloarthritis precede axial symptoms, this may be used as beginning of the "disease duration". This should however be documented. (3) The onset of signs for an associated SpA without AS-specific symptoms should not be used as beginning of the “disease duration” of AS. (4) The duration since diagnosis of AS by a healthcare provider may be of interest in some connections of research but should never be called "disease duration".

Ernst Feldtkeller, PhD
Vice president of the Ankylosing Spondylitis
International Federation
Michaeliburgstr. 15, D-81671 München, Germany
Tel: 0049-89-493376
Fax: 0049-89-49003836
feldtkeller@t-online.de

Jon Erlendsson, MD, rheumatologist
President of the Ankylosing Spondylitis International Federation
Grønningen 10, DK-8700 Horsens, Denmark
jon.erlendsson@stofanet.dk

References

1. Davis JC, Dougados M, Braun J, Sieper J, van der Heijde D, and van der Linden S for the ASAS (ASsessment of Ankylosing Spondylitis) International Working Group. The Definition of Disease Duration in Ankylosing Spondylitis: Reassessing the Concept. Ann Rheum Dis 2006 Feb 7; [Epub ahead of print]

2. Spoorenberg, A. et al., A comparative study of the usefulness of the Bath Ankylosing Spondylitis Functional Index and the Dougados Functional Index in the assessment of ankylosing spondylitis. J Rheumatol 1999;26:961–965

3. Feldtkeller E. Erkrankungsalter und Diagnoseverzögerung bei Spondylarthropathien. Z Rheumatol 1999;58:21–30

4. Feldtkeller E, Bruckel J, Khan MA. Scientific contributions of ankylosing spondylitis patient advocacy groups. Curr Opin Rheumatol 12 (2000) 239-247

5. Feldtkeller E, Khan MA, van der Heijde D, van der Linden S, Braun J. Age at disease onset and diagnosis delay in HLA-B27 negative vs. positive patients with ankylosing spondylitis. Rheumatol Int 2003;23:61–66

6. Brophy S, Mackay K, Al-Saidi A, Taylor G, Calin A. The natural history of ankylosing spondylitis as defined by radiological progression. J Rheumatol. 2002;29:1236–1243

7. Feldtkeller E, Vosse D, Geusens P, van der Linden S. Prevalence and annual incidence of vertebral fractures in patients with ankylosing spondylitis. Rheumatol Int. 2006 Jan;26(3):234-239