Generator Pocket Adhesions of Cardiac Leads: Classification and Correlation with Transvenous Lead Extraction Results HECTOR RODRIGUEZ CETINA BIEFER, M.D.,* DAVID H ¨ URLIMANN, PH.D.,† J ¨ URG GR ¨ UNENFELDER, PH.D.,* SACHA P. SALZBERG, PH.D.,* JAN STEFFEL, PH.D.,† VOLKMAR FALK, PH.D.,* and CHRISTOPH T. STARCK, M.D.* From the *Clinic of Cardiac and Vascular Surgery; and †Clinic of Cardiology, University Hospital Zurich, Zurich, Switzerland Objectives: Pacemaker (PM) and implantable cardioverter defibrillator (ICD) leads become encapsulated intravascularly and in the generator pocket by fibrotic adhesions that accumulate over time. These adhesions are responsible for the difficulty and risk of lead extraction procedures. We developed a classification scheme for pocket adhesions, classified all of the patients in the cohort, and examined the relationship between pocket adhesions and the outcome of the procedure. Methods: The classification of adhesions with respect to the intraoperative adhesion coverage was as followed: class 0 = adhesion free; class 1 ≤ 30% of adhesion coverage; class 2 = 30–60% of adhesion coverage; and class 3 ≥ 60% coverage. Patient data between December 2010 and March 2012 were collected. A total of 100 leads were extracted from 58 patients (1.7 ± 0.8 leads/patient); the mean lead implant duration was 78.5 ± 66.7 months, and the percentage of PM/ICD leads was 68% (n = 68)/32% (n = 32). Results: Distribution of the leads among classes: 0 = 10; 1 = 17; 2 = 25; and 3 = 48. Average implant times (months) according to the adhesion classes: 0 = 1.2 ± 0.4; 1 = 19.8 ± 19.2; 2 = 79.3 ± 46.6; and 3 = 115.1 ± 106.0 (correlation-coefficient 0.71; P ≤ 0.05). Average numbers of extraction tools used according to the adhesions: 0 = none; 1 = 0.4 ± 0.7; 2 = 1.6 ± 1.0; and 3 = 2.3 ± 1.2 (correlation coefficient = 0.67; P ≤ 0.05). Complete removal was achieved in 100% of the patients in classes 0 and 1; 96% in class 2 (n = 24); and 75% in class 3 (n = 36) (P ≤ 0.05). Mortality = 0. Conclusions: Extensive adhesions in the generator pocket predict the need for a higher number of extraction tools. High-grade pocket adhesions predict lower success rates with regard to complete lead extraction. Both findings suggest that the degree of pocket adhesions predicts the degree of intravascular adhesions. (PACE 2013; 36:1111–1116) surgery, lead extraction, pacing, defibrillation – ICD, classification Introduction Recent reports have shown a worldwide in- crease in the implantation of cardiac rhythm man- agement devices (CRMs), which include cardiac pacemakers (PMs) and implantable cardioverter defibrillators (ICDs), with numbers reaching 500,000 devices per year. 1–6 This increase is primarily attributed to the increasing number of Conflict of Interest: Drs. Starck and H¨ urlimann receive workshop honoraria from Cook Medical Europe Ltd. The other authors declare no conflict of interest. Address for reprints: Christoph T. Starck, Clinic of Cardiac and Vascular Surgery, University Hospital Zurich, R ¨ amistrasse 100, 8091 Zurich, Switzerland. Fax: +41 44 255 44 67; e-mail: christophthomas.starck@usz.ch Received September 3, 2012; revised March 25, 2013; accepted April 2, 2013. doi: 10.1111/pace.12184 indications for the implantation of such devices during recent years. 4,7–9 In addition to the increased number of CRM implants, an increasing number of lead extractions has been reported; these extractions are primarily caused by device-related infections, which are the leading indications for lead extractions. 1–3,10 It is generally easy to remove the pulse generator from a system. However, the extraction of implanted leads may be challenging or sometimes impossible due to the process of encapsulation by fibrotic attachments along the length of a lead, which increases over time. 11–14 The severity of fibrotic adhesions varies, and with the exception of implant duration, no factors can currently predict the extent of the adhesions that can be expected during an extraction procedure. The risks and successes of a lead extraction procedure are determined by the degree of intravascular adhesions present. 11 Therefore, it is desirable to ©2013, The Authors. Journal compilation ©2013 Wiley Periodicals, Inc. PACE, Vol. 36 September 2013 1111