Introduction: Systemic Sclerosis (SSc) manifests as inflammation, vasculopathy and fibrosis of the skin and visceral organs. Cardiac involvement, defined as either pericarditis, congestive heart failure, severe arrhythmias, and/or atrioventricular conduction abnormalities, was found to be the leading cause of death (36%) in a large cohort of SSc patients, followed over 10 years. Literature reports 4–6% of SSc patients die of sudden cardiac death (SCD). We described SSc patients with ICD implants, under the care of the Royal Free Hospital (RFH) Scleroderma Service. RFH is a national tertiary referral centre for SSc patients and the largest UK cohort.
Methods: A retrospective study was conducted on patients registered under the RFH Scleroderma Service who had ICDs implanted between 2008 and 2018. All medical notes available were reviewed. We recorded their demographics, scleroderma profile, cardiac biomarkers, NYHA class, ECG, echocardiography, cardiac MRI findings, ICD complications, ICD activations and date of death/last follow up.
Results: ICDs were implanted in 24 SSc patients between 2008 and 2018. The mean age was 48 ± 13 (range 18–74). Fifteen of the 24 patients had diffuse cutaneous systemic sclerosis (dcSSc) and nine of these had an overlap with another autoimmune condition (6 of these 9 had myositis). Of the nine limited cutaneous systemic sclerosis (lcSSc) patients, four had overlap (two of the four had myositis).
Thirteen patients had an ICD implanted for primary prevention and three of these patients went on to have appropriate shocks for VT. All three patients had an ejection fraction (EF) less than 35%. Two of these three patients had Late Gadolinium Enhancement (LGE) on their MRI. All three had NT-proBNP raised above the upper reference range of 236 ng/L.
Eleven patients had an ICD implant for secondary prevention, three had further appropriate shock therapy for VT. Two of the three patients had an EF <35%. All three had NT-proBNP less than 120 ng/L. However, two of the three had myositis and all three had LGE on MRI.
Device complications included, symptomatic subclavian thrombosis, lead dislodgement post implant and post appropriate shock therapy requiring leads to be repositioned. Three dcSSc patients had inappropriate shocks, two of these died within 12 months of implant.
Conclusions/implications: This is a small retrospective study, from a selective cohort within a single specialist centre. These limitations mean any conclusions should be thought of as hypothesis generating only. Patients requiring secondary prevention ICDs may have preserved EF and still be at risk of potentially fatal ventricular arrhythmias. LGE on MRI is present in both primary and secondary prevention patients with ventricular arrhythmias. The association between LGE on MRI and the risk of SCD needs further confirmation in larger prospective studies.
Complication rates (25%) from ICD implantation are higher than experienced in the general ICD population. Death rates within 12 months (29%) post implant (64% of total deaths) are also high, despite ICD implantation. Thus, prescription of ICDs within this group needs further risk stratification.
Clinicians looking after SSc patients need to be conscious of the risk of SCD in this group. ICD implantation does offer an opportunity to prevent SCD from arrhythmias; however, there is limited evidence to identify those who will derive benefit from ICD implantation. To further refine the risks and benefits of ICD therapy in this population future studies should include patients with myositis overlap and LGE on MRI.