Takayasu arteritis (TA) is a large vessel vasculitis of unknown aetiology affecting mainly aorta and its major branches.1 Nearly 50% of patients with TA can have cardiac involvement during their course of the disease, and is the major cause of morbidity and mortality.2 Myocardial involvement may be due to myocarditis or myocardial ischaemia secondary to coronary arteries involvement manifesting as myocardial fibrosis, with reported range from 7% to 83% of cases in different studies.3–7 Endomyocardial biopsy remains the gold standard for the diagnosis of myocarditis. However, its being invasive, relative lack of sensitivity, combined with limited availability of the experienced operators, equipment and cardiovascular pathology resources and procedure related complications limits its use.8 CMR is an excellent non-invasive modality which is more accurate and reproducible technique for assessment of cardiac morphology and function than any other method.9,10 Myocardial fibrosis,11,12 indicative of remote myocardial disease, can only be diagnosed by CMR.

Assessment of disease activity in TA is a less defined and is a challenge for the clinicians. Indian Takayasu clinical Activity Score (ITAS2010) is a composite score of disease activity developed in India, and its version incorporating serum acute phase reactants (ESR and CRP) is ITAS-A. It is better than physician global assessment (PGA) in term of comprehensiveness and the inter-rater agreement.13

In the context of limited literature on involvement of myocardium in TA, this study was planned to evaluate prevalence of subclinical myocardial involvement in TA by CMR and its correlation with disease activity score (ITAS 2010 and ITAS-A).

A total of 37 patients classified as TA,14 were included after informed consent (Fig. 1). Demographic, clinical and laboratory data were recorded in the predesigned proforma. All patients were evaluated by transthoracic echocardiography in standard parasternal, apical, and sub-xiphoidal view by an experienced cardiologist (AB). Disease activity was recorded by ITAS2010 and ITAS-A (AC).

Fig. 1.

Patient flow diagram.

(0,28MB).

CMR was done on 3.0Tesla Philips Insignia (Philips healthcare, Best, The Netherlands) system. Evaluation has been done by a single expert (MS) blinded to the clinical data on a dedicated work station (Philips Intel space portal).

The mean age of patients (n=37) was 28.92±10.7 years (range 14–58 years). There were 28 females (75.68%) and 9 males (24.32%) with a female to male ratio of 3.1:1. The mean age of symptom onset was 23.86±8.4 years (range 11–46 years). The median disease duration was 5.1±5.1 years (range 1 year to 27 years). Claudication was the commonest presenting symptoms (89%), followed by easy fatigability (73%), headache (67%), dyspnoea (57%), chest pain (32%), joint pain (19%), and syncope (16%). Details of baseline characteristics provided in Table 1.

Examination at recruitment revealed unequal pulse in 32/37 (87%) patients, bruit in 28/37 (76%), and hypertension in 24/37 (65%) at recruitment. Two patients (5%) had diabetes with disease control by oral hypoglycaemic agents. Four patients (11%) were hypothyroid, and were under control by oral levothyroxine.

Both left subclavian, and descending thoracic aorta were affected in 28 (75%) patients, the commonest vessels involved in the cohort. The abdominal aorta was the second most common, in 26 (70%) patients. The cohort was classified into six groups as per Numano's angiographic types. Type V was the commonest type present in 23 (62%) patients followed by type III in six (16%), type I in four (10%), type II (combined IIa and IIb) in 3 (8%) and type IV in one (2%) (Fig. 2).

Fig. 2.

Distribution of angiographic types of Takayasu arteritis patients.

(0,13MB).

Myocardial involvement on CMR was noted in five patients (n=5/37). Two patients (2/5) had CMR features of myocarditis (Figs. 3 and 4). Three (3/5) had CMR evidence of ischaemic myocardial fibrosis. Details of myocardial involvement and distribution are provided in Table 2.

Figs. 3 and 4.

(3A and 3B): T2-STIR (short tau inversion recovery) cardiac magnetic resonance imaging (CMR) images in horizontal (A) long axis show myocardial hyperintense signal in the apex (arrows). Late gadolinium enhancement (LGE) apical short axis image shows confluent epicardial predominant myocardial enhancement. On cine cardiac MRI imaging only subtle hypokinesis was seen (not shown here). (4A–4C): CMR, LGE images in short axis at base (A), mid cavity (B) and at apex (C) shows transmural enhancement in the anterior segments (thick arrow in A and B) and sub-endocardial in lateral and inferior segments (A and B) and at apex (C) in the left anterior descending and left circumflex territories (7/17 segments).

(0,12MB).

On the evaluation, there were higher percentage of patients in the age group of <18 years, higher CRP value, more patients with active disease (both ITAS2010 and ITAS-A), and less number with conventional risk factors like high cholesterol, LDL, and Framingham risk score (FRHS) although none of those parameters had achieved statistical significance level (Table 3).

Relationship with arterial bed involvement: significant association with right renal artery involvement (p=0.037) and higher but non-significant association with the abdominal aorta (p=0.085) and left renal artery (p=0.079) involvement.

Myocardium involvement has been reported in TA3–7; however, there are limited studies on CMR evaluating myocardial involvement in TA.8,9 Myocardium can be affected as myocarditis, ischaemia (due to coronary artery involvement), secondary to valvular disease or reno-vascular hypertension. Up to 50% of patients of TA can have subclinical myocardial inflammation as documented by different histopathological studies.4,17 Endomyocardial biopsy though remains the gold standard, however, CMR is now the modality of choice due to non-invasive imaging without any hazard of ionizing radiation. It also offers better anatomical visualization, less inter-observer variability; more accurate quantitative assessment and tissue characterization based on pattern of myocardial enhancement on late gadolinium enhancement (LGE) images.

Our study enrolled 37 patients with mean age of 28.9±10.7 years compared to 49±14 and 48 (34–61) years, in previous studies on cardiac involvement in TA, evaluated by CMR.5,6 The current study showed a changing trend towards female dominance compared to the previous studies.15 Initial presentation with systemic symptoms like fever and weight loss were a distinct finding compared to the previous studies. Claudication pain (89%) and constitutional symptoms like easy fatigability (70%), weight loss (54%), and fever (51%) were the main symptoms in the current study, whereas systemic symptoms were present only in 16% of cases and hypertension was the commonest in previous Indian studies.16 According to the distribution of vessels type V was the most common variant, consistent with the previous Indian data. But there are substantial changes in the distribution of vessel involvement. In the current study, there is significant proximal migration of vessel involvement compared to previous data.

In our study protocol myocardial involvement was characterized as due to myocarditis, ischaemia or non-specific changes. Two previously reported studies on the same subject reported myocardial involvement but were silent on further characterization.5,6

In current study 14% (5/37) had myocardial involvement as detected by CMR with 2/5 having myocarditis and 3/5 having features of ischaemic myocardial fibrosis secondary to coronary arteries involvement. CMR diagnosed myocardial involvement in two cases (Table 2, P1 and P5), which were missed by echocardiography.

In 13.5% (5/37) patients in our study had CMR abnormality compared to 27% and 26% in the previous studies (Table 4).5,6 Of the three patients, 2 (5%) had LGE displayed typical pattern of myocardial ischaemia (MI), whereas study by Comarmond et al.6 in 22% had LGE distribution typical of MI, and had three times higher association with renovascular hypertension. Compared to the previous studies the current study had lower age of onset, less disease duration and lower number of diabetic patients. These factors may contribute to higher percentage of cardiac MRI abnormalities in previous studies. LGE typical of myocardial fibrosis in CMR can be found in many condition including hypertension.18 The main difference from the earlier studies were mean age, age of onset, and disease duration, all of which were much less in the current study. Other than genetic and ethnic variation, the higher percentage of LGE detected by the previous studies may be a reflection of cumulative damage with increasing age, prolong hypertension and disease duration.

We recognize various limitations of this study. The optimal timing for CMR in TA and effect of treatment on CMR findings has not been addressed. Published literature till date has also not commented on this valid issue. Further large multi-centric studies are thus required for specific answers. Another limitation was that not all the patients had undergone coronary angiography for assessment of coronary arteries.

Prior presentation: The data was presented partially at EULAR e-Congress 2020, as a poster presentation.

This was the largest cohort of TA evaluated by CMR showed prevalence of subclinical myocardial involvement was much less (8% vs. 25–27%) compared to the previous studies. Myocarditis was not mentioned in the previous studies. Myocardial involvement trend towards early age of onset, less disease duration, lack of classical risk factors, and more with disease activity.

The current study showed a change in demography, presenting clinical features, and distribution of arterial bed involvement in Indian patients with TA.

Echocardiography may miss some cases of myocardial involvement, especially the acute myocarditis which can only be detected by CMR, hence it can miss the window of opportunity of treatment. Judicious use of CMR may help in detecting subclinical myocardial involvement and can alter the treatment outcome.

None.

None of the study authors have any conflict of interest.