Results
Baseline Characteristics
The mean age of the study population at presentation was 43.33±13 years with a male predominance (73.3%). The left ventricular ejection fraction at presentation was 48.06±12.29% and the left atrial volume was 37.6±10.64 ml. None of the patients had the clinical features of acute rheumatic fever, tuberculosis, or systemic sarcoidosis. Baseline clinical characteristics are given in table 1. One of the patients had a Pacemaker for sinus nodal dysfunction. Median follow up was 30.53±13.55 months. None of the patients had ventricular arrhythmias at presentation. At presentation, the NYHA class was 3±0.5. Four patients (26.67%) presented with an ischemic stroke and one patient had a left atrial thrombus (6.7%). Spontaneous echo contrast was noted in 3 of the 15 patients (20%). The mean CHA2DS2 Vasc score of the patients at presentation was 1.06±0.93.
Atrial arrhythmias
Atrial fibrillation was the presenting arrhythmia in all these patients and atrial flutter (AFL) was observed in 4 patients (26.7%). Two patients (13.3%) also had typical atrioventricular nodal re-entrant tachycardia (AVNRT). One of the patients presented with AF, AFL and an AVNRT and 5 patients had another dysrhythmia in addition to AF. The AF was paroxysmal in 8 patients (53.3%) and persistent in 7 patients (46.7%). The patients had significant symptoms such as dyspnoea (85.7%), chest discomfort (57.1%), palpitations (78.6%) and light-headedness (50%). None of the patients had syncope. Four of these patients presented with stroke (26.67%) during which AF was detected. All the strokes were ischemic and none of the patients had haemorrhagic stroke. Cardioversion was performed in 9 of these patients (60%). At follow up, 4 (26.7%) of these patients had recurrent AF episodes.
Imaging
18F FDG-PET scans showed evidence of atrial inflammation in all patients. The atrial uptake was seen in the right atrial appendage in 7 patients (46.7%), right atrial wall in 13 patients (86.7%), interatrial septum in 2 patients (13.3%), left atrial appendage in 4 patients (26.7%), anterior left atrial wall in 5 patients (33.3%), and posterior atrial wall in 4 patients (26.7%). (Table 2) Biatrial involvement was observed in 5 patients (33.3%). In addition, increased uptake in the left ventricular myocardium was noted in 3 patients (20%). This was noted in the basal septal region in 2 patients and at the LV apex in 2 patients and LV lateral wall in one patient.
A CMR was performed in 6 patients and was normal in 5 patients. One patient had evidence of delayed enhancement in the midmyocardial aspect of the anterolateral LV. No atrial abnormalities were detected on MRI.
Extracardiac involvement
Lymphadenopathy was the only extracardiac involvement and this was observed in the mediastinum in 12 patients (80%), axillary region in 1 patient (6.7%), cervical region in 2 (13.3%) and supraclavicular region in 1 patient (6.7%). 18F FDG PET showed uptake in mediastinal nodes in 15 patients (100%) and in other regions in 3 patients (20%). (Table 2) The HRCT of the chest showed no abnormalities in the pulmonary parenchyma. There was no clinical evidence of rheumatological disease.
Histopathology
Eleven of the patients underwent needle aspiration of the mediastinal lymph nodes, while the remaining patients had lymph node biopsy from cervical, supraclavicular, or axillary regions. One patient had biopsy of the RA appendage ( surgical ) . There was evidence of inflammation in all patients either on endomyocardial biopsy or biopsy of lymph nodes. 14 (93.3%) patients showed evidence of granulomatous inflammation. Granulomatous inflammation without caseation was noted in 11 patients (73.3%) while caseation or necrosis was observed in 3 (20%) patients. One patient showed evidence of lymphocytic infiltrate suggestive of lymphocytic myocarditis.
Underlying disease
The tuberculin skin test was positive in 5 patients (33.3%) out of these 3 patients showed evidence of caseation and necrosis in biopsy specimens. M. Tuberculosis deoxyribonucleic acid PCR was positive in 1 of the 15 patients (6.7%) and Mycobacterium tuberculosis was cultured in two patients on lymph node biopsy sample (13.3%). None of these patients exhibited evidence of pulmonary tuberculosis either at presentation or anytime in the past. So, Tuberculosis was the underlying aetiology of granulomatous inflammation in 3 patients (20%).
Sarcoidosis was diagnosed based on the Expert Consensus Diagnostic criteria in the remaining 12 patients (80%).
Electrophysiological study (EPS)
Out of these 15 patients, an EPS with radiofrequency (RF) ablation of atrial flutter was performed in 3 of these patients (20%). Two patients received a cavotricuspid isthmus line for right atrial flutter, one patient had an ablation of left atrial flutter and one patient had a slow pathway modification for a typical AVNRT. The RFA was successful in 3 of these patients and unsuccessful in one patient.
Management
All patients received disease specific therapy in the form of immunosuppression or anti Tuberculosis therapy. Immunosuppression therapy included Prednisolone or methotrexate. Patients received rate control with beta blockers, calcium channel antagonists and digoxin. Antiarrhythmic drugs used for rhythm control were flecainide, amiodarone and sotalol. The clinical characteristics and therapies are summarized in Table 2. RFA of atrial flutter and supraventricular tachycardia was performed at the onset in 3 patients along with disease specific therapy. One patient who had an unsuccessful ablation and was treated with disease specific therapy and had no recurrence of arrhythmia.
Overall, the initiation of disease specific therapy reduced the incidence of atrial arrhythmias at 30.53±13.55 months of follow up. This corresponded to resolution of atrial inflammation on 18F FDG PET-CTs. There was a significant improvement in functional class of patients from NYHA 3±0.5 To NYHA 1.2±0.2 at follow up (P=0.005). Importantly, there was an improvement in LV ejection fraction from 48.07±12.29% to 56±12.07% but this did not achieve statistical significance (P =0.0853). The RA and LA volumes were not significantly different from baseline. There was a significant decline in inflammatory parameters, with the hs-CRP declining from a baseline value of 11±11.94 to 2.9±2.3 at follow up (P=0.0215). Recurrence of atrial arrhythmias was noted in 4 patients at follow up and these two of the patients needed another cardioversion and escalation of immunosuppression due to persistent 18F FDG PET uptake.
One patient had a decline in LVEF at follow up and had Ventricular uptake on 18FDG PET-CT.
Oral Anticoagulation
14 out of the 15 patients were treated with oral anticoagulation with vitamin K antagonists to maintain an INR of 2-3. One patient was not anticoagulated based on patient preference.