Management of isolated device-related infective endocarditis: a nationwide study from the NIDUS registry

Abstract Background Current guidelines recommend the removal of cardiac implantable electronic devices (CIEDs) in patients with CIED-related infective endocarditis (IE). However, in some cases, patients are managed without CIED removal, and it remains uncertain whether omitting CIED removal is safe. Purpose We aimed to assess patient characteristics, management, and outcomes in patients with isolated CIED-related IE, stratified by CIED removal status. Methods This nationwide study included all patients with first-time isolated CIED-related IE from 2016 to 2021, identified using the NatIonal Danish Endocarditis StUdieS (NIDUS) registry. The outcomes, relapse of IE or bacteraemia, each caused by the same microorganism, were assessed within six months after IE discharge. We used multivariable logistic regression to calculate an odds ratio (ORs) for in-hospital mortality comparing CIED removal to no removal, adjusted for sex, age, CIED type, diabetes, heart failure, stroke, and chronic obstructive pulmonary disease. Results We included 355 patients with isolated CIED-related IE, of whom 278 (78.3%) underwent CIED removal. Patients without removal were significantly older (median age: 80.7 years IQR: 74.2-85.6 vs. 74.6 years IQR: 66.5-79.9). The distribution of CIED types was similar between the two groups, with 56.6% having pacemakers. Likewise, the prevalence of blood culture-negative IE, staphylococci, streptococci, and enterococci, as well as comorbidities, was comparable between the two groups. A lower proportion of patients without CIED removal were self-reliant (46.8% vs. 72.6%), and a higher proportion were from nursing homes (29.3% vs. 5.6%). Sepsis at admission was present in 27.3% of the non-removal-group and 18.7% of the removal group (P=0.10) (Figure). The length of hospital stay did not differ significantly between the non-removal and removal groups (33 days IQR: 18-48 vs. 35 days IQR: 25-44). In-hospital mortality was 20.8% in the non-removal group and 9.0% in the removal group (P=0.004). Among patients without CIED removal, 19.5% were discharged alive on chronic suppressive antibiotic therapy (Figure), while 59.7% were discharged alive without such therapy. Overall, relapse of IE occurred in 7 (2.0%) and relapse of bacteraemia in 8 (2.3%), with a similar prevalence observed between the two groups. In adjusted analysis, CIED removal was associated with lower in-hospital mortality (OR: 0.43 95% CI: 0.21-0.91). Conclusions In this nationwide study, most patients with isolated CIED-related IE underwent CIED removal as recommended by guidelines. Patients who did not undergo CIED removal were older, more frail, and had higher in-hospital mortality. Among these patients, 60% were discharged alive without chronic suppressive antibiotic therapy. Relapse of either IE or bacteraemia was comparable between the non-removal and removal groups. Individualised management strategies remain essential for improving outcomes.