Analysis of Outcomes in 8304 Patients Undergoing Lead Extraction for Infection

Background Patients undergoing lead extraction for infected devices have worse outcomes compared with those with noninfected devices. We assessed predictors of in-hospital mortality and procedure-related major adverse events (MAEs) in a large cohort undergoing lead extraction. Methods and Results Deidentified hospital records procedure from 7 states between 1994 and 2013 were aggregated and International Classification of Disease, Ninth Revision (ICD-9) procedure codes were used to identify hospital records reporting lead extraction. MAEs included death, cardiac tamponade, hemothorax, and need for emergent cardiac surgery. Predictors of in-hospital MAEs for infected compared with noninfected leads were identified using multivariate regression. Associations between outcomes and specific microbe were also assessed. In total, 57 220 discharges specified lead extraction. Infected leads accounted for the minority of total lead extractions compared with fractured leads (16.1 versus 59.8%, 25.7% not reported). There were 3298 MAEs (5.8%) including 980 deaths (1.7%). Multivariate predictors of MAE included black race, atrial fibrillation, anemia, heart failure, and admission via either hospital transfer or emergency department versus home (all P<0.001). Infected leads were associated with an increased risk of death (4.6% versus 0.9%, P<0.001) compared with leads with fracture only. Among patients with microbial data, staphylococcal infection was most common, whereas streptococcal infection was associated with the worst outcomes. Conclusions Patients undergoing extraction of infected leads have higher in-hospital mortality and adverse events compared with noninfected leads. Streptococcus, anemia, and heart failure are predictors of adverse outcomes.