Infection is a potentially catastrophic consequence of arrhythmia device surgery, since it is usually requires device removal and lead extraction. Infection risk is increased in patients undergoing replacement or receiving a complete de novo device, and is reduced by use of preoperative intravenous antibiotics. Incremental interventions to reduce infection are often performed on an ad hoc basis in the absence of informative clinical trial evidence or specific guidelines.
Eleven sites enrolled high-risk patients (device replacement, revision or CRT) in an unblinded pilot study to establish feasibility and enrollment rate for a definitive randomized trial. Patients were randomized to single-dose preoperative iv antibiotics (cefazolin, clindamycin or vancomycin) versus single-dose iv antibiotics plus intraoperative pocket-wash (bacitracin or vancomycin) plus a 3-day post-operative course of oral cephalexin or clindamycin.
Over a 12-month period, 500 patients were enrolled (age 67±14 years, 70% male). The majority of patients underwent replacement (71%) or revision/upgrade (28%). Diabetes was present in 30%, 54% were taking ASA and 82% of procedures were performed by electrophysiologists (82%). After a one-year follow-up, proven or suspected device infection occurred in 9 of 249 patients randomized to conventional therapy vs. 4 of 251 patients randomized to incremental antibiotics (3.6 vs. 1.6%, p=0.17). Proven infection leading to hospitalization was present in 5 conventional patients vs. 3 incremental patients (2.0 vs. 1.2%, p=0.50). These event rates were consistent with the pre-study estimated event rates and effect size. System extraction was performed in 3 conventional patients and no patients randomized to incremental antibiotics. Antibiotic related adverse events were noted in 3 conventional and 5 incremental patients (1.2 vs. 2.0%, p=0.50).
The rates of device infection in the current pilot are consistent with published estimates from registries and previous studies, supporting the need for a definitive randomized trial. Such a trial would require comparison of at least 5000 patients in each arm to detect a 35% difference in event rate with 80% power. The PADIT study team is now embarking on a definitive randomized trial supported by peer-reviewed funding from the Canadian Institute of Health Research.
© 2012 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.