A single centre retrospective analysis of the management and outcomes of cardiac device-related infective endocarditis

Abstract Background/intro Cardiac device-related infective endocarditis (CDRIE) is a rare but serious complication of cardiac device insertion with a high mortality rate. Cardiac devices are increasingly being implanted in older patients amid an ageing population with more comorbidities. Our centre is a large cardiac surgery referral centre in London. The centre implants a large number of cardiac devices per year (an average of 2168 per year from 2016 - 2019). The IE multidisciplinary team (IE MDT) manages approximately 150 IE referrals from the region per year. Purpose We aimed to describe a cohort of patients with CRDIE, referred to the IE MDT over 5 years (2015-2020), to generate a contemporary representation of patients with CDRIE. Methods We carried out a retrospective review of all referrals to the IE MDT between 2015 and 2020, identifying all patients with CDRIE - infection extending to the electrode leads, cardiac valve lead(s) or endocardial surface (ESC). All patients who were discussed at the IE MDT between October 2015 and March 2020 were screened, based on inclusion criteria according to the ESC's definition of CDRIE. Patients who had an infection of the cardiac device pocket with no evidence of IE were excluded. Results 60 out of 928 IE patients had confirmed CRDIE after discussion in the IE MDT from 10/2015 and 03/2020. The average age of the cohort was 70.3 (male =44, 73%). 61.7% of the devices were PPM (n=37), 10% were ICD, 20% were CRT-D, and 8.3% were CRT-P. The average duration of device therapy before removal was 64 months. 66.7% of devices were extracted (n=40) and 33.3% were managed medically (n=20). 43.3% of the cohort had prosthetic valves (n=26), and 13.3% of the total cohort had tissue aortic valve implants (TAVI) (n=8). 85% had a transesophageal echo (TOE) procedure (n=51) and 61.7% went on to have a PET CT scan (n=37). The overall inpatient mortality was 25% (n=15), rising to 45% at 12 months (n=27). Patients managed medically had a significantly higher mortality at 6 and 12 months compared to those who received an extraction and antibiotics (figure 1). There was no difference in left or right ventricular systolic function pre- and post- treatment on average or degree of tricuspid regurgitation. Medically managed patients were frailer by Charlson Index (average Charlson Index score 3) than surgically managed patients (average Charlson Index score 2). Conclusions In a retrospective assessment of CRDIE, we found that there is a low incidence of CRDIE complicating cardiac device insertion at BHC when compared to the national average. We found that surgical removal improves the likelihood of survival at 6 and 12 months when compared to medical management in those with a severe Charlson score, suggesting that even in cases of frailty and multiple morbidities, device extraction is preferable (figure 2).Survival curve of all CDRIE patientsSurvival curve in the severely frail