LAUSR

Dear Editor, With great interest, we read the original article ‘Detection and evaluation of haemophilic arthropathy: Which tools may be considered more reliable’ by Plut et al.1 The authors assessed the correlation of ultrasound (US), clinical examination by the Haemophilia Joint Health Score (HJHS), patient's subjective assessment and certain laboratory biomarkers with magnetic resonance imaging (MRI) in order to determine which tool is most reliable. They conclude that only US should be included into the regular followup of patients with severe haemophilia. The investigated biomarkers s25OH vitamin D, sferritin, sCterminal telopeptide of type I collagen, sNterminal propeptide of type I procollagen and scartilage oligomeric matrix protein showed no correlation with the presence or degree of haemophilic arthropathy, and as such the authors concluded that these biomarkers may not be useful for the (early) detection and evaluation of haemophilic arthropathy. We agree that biochemical markers in haemophilic arthropathy are still in an experimental stage and not yet sufficient for implementation in daily practice. However, as biomarker research is challenging, we would like to emphasize some caveats in this field. Currently, the progression of haemophilic arthropathy is monitored by clinical examination and the use of imaging techniques. These methods only provide information on past dynamic processes and have their specific limitations. We and many other researchers advocate to develop a tool reflecting minor changes in an early stage of arthropathy in order to detect progression even before it results in clinical symptoms or changes on imaging. Inflammatory or joint tissue (bone, cartilage and synovium) turnover markers may provide this significant information and can potentially evaluate the joint status more accurately.2 Laboratory biochemical markers reflect the total joint score and therefore the presence or degree of arthropathy at individual joint level is not representative. Moreover, one must consider how to weigh prosthetic joints in this respect. Although it represents ‘endstage’ severity of arthropathy, these joints will not contribute to the release of cartilage markers. However, bone markers will be released, and in the evaluation of such markers, prosthetic joints should be scored as the worst joint condition. Recurrent joint bleedings induce both inflammatory and degenerative changes. Some patients suffer from chronic synovitis without evident osteochondral damage, while other patients develop significant cartilage and bone changes without clear synovitis. As such, joint bleeds can lead to different longterm joint conditions and also the degree of arthropathy varies in patients with a similar number of bleeds.3 This complicates research for correlations significantly and may require a different set of biomarkers for different conditions. We suggest to evaluate the correlation of biomarkers with subscores of the International Prophylaxis Study Group (IPSG) score. The IPSG score exists of a subscore for softtissue changes and osteochondral changes, respectively. This may provide better methodology to distinguish and study the role of biomarkers in relation to severity. Combined indexes of biochemical markers capturing all pathogenic processes may be an alternative. The authors suggest a comparison with a control group to enable additional evaluation of biochemical markers. In our opinion, investigating biomarkers in a (healthy) control group is not clinically relevant as differentiation between patients with haemophilia and healthy controls is performed based on factor VIII/IX activity. Another point to consider is the high interindividual difference which is a challenge in biochemical marker research. Longitudinal studies with multiple measurements within one patient to diminish variability seem more appropriate to apply. Differences in bone markers can partially be explained by different ages. In young patients, skeletal growth and puberty lead to significant changes in levels of bone markers.4 Therefore, age should be taken into account when comparing bone markers in different studies. Finally, biochemical markers are influenced by other factors present in the haemophilia population. Therefore, it is essential to include data about the most recent joint bleeding and about coinfections with human immunodeficiency virus (HIV) or hepatitis C virus (HCV). We appreciate the effort to determine biochemical markers and considering it as a tool for the detection and evaluation of haemophilic arthropathy. Although research in this area comes with challenges, we do believe that future research is important and valuable in order to develop a tool reflecting minor changes in an early stage of arthropathy. In clinical practice and trials, these markers may provide early and complementary information about the joint status. We advocate for harmonization in biomarker research, longitudinal designs to eliminate interindividual differences and welldefined joint outcome assessments.