Article Text

Download PDFPDF

Cardiac involvement in ankylosing spondylitis. Can new magnetic resonance indices interpret cardiac pathophysiology beyond echocardiography?
  1. Sophie Mavrogeni1,
  2. Genovefa Kolovou1,
  3. George Kitas2
  1. 1Onassis Cardiac Surgery Center, Athens, Greece
  2. 2Arthritis Research UK Epidemiology Unit, Manchester University, Manchester, UK
  1. Correspondence to Dr Sophie Mavrogeni, MD FESC, 50 Esperou Street, 175-61 P.Faliro, Athens 17561, Greece; soma13{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Ankylosing spondylitis (AS) associates with various cardiac lesions, including aortitis/aortic regurgitation, conduction abnormalities and myocardial involvement, due to either coronary artery disease or primary cardiomyopathy.1 The cornerstone tool for the non-invasive evaluation of heart involvement has, until recently, been echocardiography.1 Echocardiographic evaluation in AS is mainly focused on valve and ventricular function assessment, with the pattern of ‘diastolic dysfunction with preserved ejection fraction’ being the most common preclinical finding. A strong correlation has also been shown between several indices of vascular pathology associated with the development of cardiovascular disease (CVD) and increased disease activity in patients with AS.2

Cardiovascular magnetic resonance (CMR), a non-invasive, non-radiating, operator-independent technique, currently represents the gold standard for ventricular function and tissue characterisation in patients with CVD, including those in whom CVD develops in the context of a connective tissue disease (CTD).3 ,4 Recently, new CMR indices, such as native and postcontrast T1 mapping, T2 mapping and extracellular volume quantification (ECV) have been successfully used to identify preclinical cardiac lesions early on, in patients with rheumatoid arthritis (RA).5

In the Heart study by, Biesbroek et al6 used T1, T2 mapping and ECV to allow quantitative characterisation of myocardium, beyond the conventional CMR approach in a small population of patients with AS. Native T1 changes can detect various pathological processes related to oedema, protein, lipid and iron deposition (haemorrhage, siderosis), without the use of gadolinium: this has serious clinical implications particularly in patients with reduced glomerular filtration rate.5 The detection of oedema is very important, as it reflects acute myocardial involvement in various CTDs, including AS.3 In the study of Biesbroek et al, although native T1 and T2 mapping, representing myocardial oedema, were higher with increasing disease activity, their correlation was not statistically significant. One potential explanation might be the small patient sample. However, another potential explanation might be that the imaging pattern of myocardial oedema does not necessarily follow the disease activity pattern, as expressed by the disease activity indices. Recently published experience in other diseases such as iron overload in thalassaemia syndromes suggests that the heart presents a rather ‘distinct pattern’, which is not always in agreement with the status of the total body iron loading.7 In CTDs, myocardial involvement has already been identified in RA patients with clinically quiescent systemic disease, as a first sign of disease relapse.8 The exact pathogenic processes, temporal association and resulting treatment requirements between myocardial involvement and other characteristics of disease activity in various CTDs require further investigation.

Late gadolinium enhancement (LGE), the gold standard for fibrosis detection, did not correlate with disease activity. This is due to the fact that LGE shows only differences between affected and non-affected tissue and fails to detect diffuse myocardial disease. In contrast, ECV, an important new CMR index sensitive to diffuse myocardial fibrosis, was strongly correlated with inflammatory indices and could discriminate patients at different stages of the disease. Early data indicated that ECV is of equal importance with left ventricular ejection fraction (LVEF) in terms of prognosis; however, LVEF underscores the biological significance of the interstitium, which can be reliably evaluated by ECV. Although the clinical significance of extracellular matrix in various diseases still remains under evaluation, it has been already proved useful for the detection of silent diffuse fibrosis in various CTDs.4 ,8 Furthermore, the diffuse distribution of collagen is impossible to be detected by LGE and therefore, the association of heart disease with outcomes in various inflammatory cardiomyopathies appears stronger for ECV than with LGE.4 ,9

It seems that the new, sophisticated magnetic resonance indices may influence the classic clinical approach of heart disease assessment in patients with AS. These indices promise to improve both diagnosis and prognosis. However, several factors, such as magnetisation transfer, diffusion distance and time, contrast mechanisms, transcytolemmal water exchange rate, flow, T2 or T2* relaxation may significantly influence the accuracy and precision of T1 measures.5 Additionally, various disease-related factors such as myocardial thickness or rhythm disturbances can potentially influence T1 mapping and introduce unknown degrees of bias.5 Furthermore, availability, expertise, cost and standardisation of these new indices in different settings should be seriously considered before their application can be recommended in routine clinical practice.

In summary, (A) although echocardiography remains the mainstream diagnostic tool for cardiac evaluation, it is unable to offer a tissue characterisation evaluation of the myocardial status in AS. (B) The classic CMR protocol including function, oedema and fibrosis characterisation can offer additional information; however, it may miss diffuse myocardial fibrosis due to LGE technical limitations. (C) The new CMR indices present better correlation with the disease activity index in AS and can provide a unique myocardial map that may facilitate personalised treatment. (D) Although new CMR indices offer great promise in the assessment of cardiac disease in AS, further evaluation is needed, before they can be recommended as tool for the diagnosis, risk stratification and treatment of cardiac involvement in AS.



  • Contributors All authors have equally contributed to this submission.

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Linked Articles