Comparative Outcomes of Extraction of Defibrillator Leads
One hundred and ninety-two patients underwent transvenous extraction of Fidelis® (n = 145) or Riata (n = 47) leads at Vanderbilt Heart and Vascular Institute from July 2006 to April 2013. Baseline demographics in the 2 groups are shown on Table 1. All Fidelis® leads extracted were model 6949. The model numbers for extracted Riata leads are listed in Table 2. A similar percentage of patients in each group underwent extraction for pocket infection (Fidelis® 6.2%, Riata 8.5%, P = 0.52), venous stenosis (Fidelis® 2.1%, Riata 0%, P = 0.32), and abandoned leads (Fidelis® 9%, Riata 10.6%, P = 0.73). Lead malfunction was a significantly more common indication in the Fidelis® group (Fidelis® 78.6%, Riata 61.7%, P = 0.02), while device-related endocarditis was a significantly more common indication in the Riata group (Fidelis® 5.5%, Riata 21.7%, P = 0.001). Twenty-one (45%) Riata leads had evidence of conductor externalization confirmed at the time of extraction. The Riata ICD lead group had a significantly longer duration since implant (P = 0.001). More leads were concomitantly extracted during procedures involving the Riata leads (median Fidelis® 1, Riata 2, P = 0.002). The excimer laser sheath was used in most patients with no significant difference between the groups (Fidelis® 93.8%, Riata 95.7%, P = 0.62). However, there was a larger median laser sheath size used for Riata leads (Fidelis® 12 Fr, Riata 16 Fr, P<0.001). There were no statistically significant differences in median procedure duration (Fidelis® 163 minutes, Riata 175 minutes, P = 0.11), or procedural success (Fidelis® 97.9%, Riata 95.7%, P = 0.41). There were 3 extractions in the Fidelis® group that were partially successful. In all 3 cases, the Fidelis® lead itself was extracted successfully at the index procedure, but the procedure was classified as a failure due to the inability to extract other in situ leads requiring removal. Two of these patients had systemic infection and complete extraction was accomplished with staged minimally invasive right thoracotomy. The third patient was free of systemic infection and clinical success was achieved with capping (abandonment) of an existing lead initially targeted for extraction. The Riata group had 2 partially successful extractions. Both patients had systemic infection and the Riata lead was successfully extracted via transvenous approach, but surgical extraction was required for older co-existing leads.
There was no statistically significant difference in the pooled complication rate between the 2 lead types (Fidelis® 5.5%, Riata 10.6%, P = 0.23, Table 3). The Fidelis® group had 8 complications. Two major complications (1.4% major complication rate) included a cardiac avulsion requiring surgery and one in-hospital death. The first patient developed cardiac tamponade immediately after the lead extraction; pericardiocentesis was performed without improvement in hemodynamics. Sternotomy was performed with successful repair of a single bleeding site in the RV. The patient remained stable and was discharged home 4 days after the procedure. The second patient presented with septic shock with evidence of lead associated endocarditis. Lead extraction was successful with no intraprocedural complications. However, she remained persistently septicemic with evidence of worsening end organ dysfunction. The family elected to withdraw life support measures on hospital day 4 due to her poor prognosis. The 6 minor complications included pocket hematomas, upper extremity swelling and need for blood transfusion. The Riata group had 5 complications including a major complication related to migration of a lead-associated thrombus to the main pulmonary artery. This patient remained hemodynamically stable and clot retrieval was performed the next day using a percutaneous snare technique. This patient was initially offered surgical extraction but a transvenous extraction was undertaken due to his unwillingness for an elective cardiac surgery. The 4 minor complications included pocket hematomas and need for blood transfusion. There was no statistically significant difference in median length of hospital stay or 30-day mortality (Fidelis® 2.1%, Riata 2.1%, P = 0.98).
Due to the higher prevalence of device infection in the Riata group, an analysis was performed excluding all patients with cardiac device-related infection. Results were similar to the primary analysis, with no statistically significant differences observed in procedure duration, procedural success, length of stay, complications, or 30-day mortality (Table 4).
Procedural characteristics and complications were compared among patients with Riata lead extractions by degree of conductor externalization. Twenty seven Riata patients had types 0 or 1 externalization and 20 had types 2 or 3. Patients with type 2 or 3 externalizations required the use of larger laser sheaths (14 Fr for type 0/1 vs. 16 Fr for type 2/3, P = 0.02). There were no statistically significant differences in number of sheaths used, procedural duration, procedural success, complications, length of hospital stay, or 30-day mortality (Table 5). Riata patients with type 2 or 3 externalization (n = 20) were compared with Riata and Fidelis® patients with no or minor (type 0 or 1) externalization (n = 172). Extractions with types 2 or 3 externalization required the use of larger laser sheaths (12 Fr for 0/1 vs 16 Fr for 2/3, P<0.001). There were no significant differences in number of sheaths used, procedural duration, procedural success, complications, length of stay, or 30-day mortality (Table 6).
Extractions of either lead type were compared based on the presence or absence of electrical lead failure (n = 122 vs. n = 70, respectively). Extractions of leads without electrical failure required larger laser sheaths (14 Fr vs 12 Fr, P<0.001), had a lower rate of procedural success (94.3% vs 99.2%, P = 0.04), a higher incidence of complications (11.4% vs 4.1%, P = 0.05), longer hospital stays (2 days vs 1 day, P = 0.001), and higher 30-day mortality (5.7% vs 0%, P = 0.008, Table 6). Leads without electrical failure were implanted for a significantly longer duration: 71% of leads without electrical failure had been implanted for more than 5 years, compared with 49.2% of leads with electrical failure (P = 0.001).
Results
Patient and Procedural Characteristics
One hundred and ninety-two patients underwent transvenous extraction of Fidelis® (n = 145) or Riata (n = 47) leads at Vanderbilt Heart and Vascular Institute from July 2006 to April 2013. Baseline demographics in the 2 groups are shown on Table 1. All Fidelis® leads extracted were model 6949. The model numbers for extracted Riata leads are listed in Table 2. A similar percentage of patients in each group underwent extraction for pocket infection (Fidelis® 6.2%, Riata 8.5%, P = 0.52), venous stenosis (Fidelis® 2.1%, Riata 0%, P = 0.32), and abandoned leads (Fidelis® 9%, Riata 10.6%, P = 0.73). Lead malfunction was a significantly more common indication in the Fidelis® group (Fidelis® 78.6%, Riata 61.7%, P = 0.02), while device-related endocarditis was a significantly more common indication in the Riata group (Fidelis® 5.5%, Riata 21.7%, P = 0.001). Twenty-one (45%) Riata leads had evidence of conductor externalization confirmed at the time of extraction. The Riata ICD lead group had a significantly longer duration since implant (P = 0.001). More leads were concomitantly extracted during procedures involving the Riata leads (median Fidelis® 1, Riata 2, P = 0.002). The excimer laser sheath was used in most patients with no significant difference between the groups (Fidelis® 93.8%, Riata 95.7%, P = 0.62). However, there was a larger median laser sheath size used for Riata leads (Fidelis® 12 Fr, Riata 16 Fr, P<0.001). There were no statistically significant differences in median procedure duration (Fidelis® 163 minutes, Riata 175 minutes, P = 0.11), or procedural success (Fidelis® 97.9%, Riata 95.7%, P = 0.41). There were 3 extractions in the Fidelis® group that were partially successful. In all 3 cases, the Fidelis® lead itself was extracted successfully at the index procedure, but the procedure was classified as a failure due to the inability to extract other in situ leads requiring removal. Two of these patients had systemic infection and complete extraction was accomplished with staged minimally invasive right thoracotomy. The third patient was free of systemic infection and clinical success was achieved with capping (abandonment) of an existing lead initially targeted for extraction. The Riata group had 2 partially successful extractions. Both patients had systemic infection and the Riata lead was successfully extracted via transvenous approach, but surgical extraction was required for older co-existing leads.
Complications
There was no statistically significant difference in the pooled complication rate between the 2 lead types (Fidelis® 5.5%, Riata 10.6%, P = 0.23, Table 3). The Fidelis® group had 8 complications. Two major complications (1.4% major complication rate) included a cardiac avulsion requiring surgery and one in-hospital death. The first patient developed cardiac tamponade immediately after the lead extraction; pericardiocentesis was performed without improvement in hemodynamics. Sternotomy was performed with successful repair of a single bleeding site in the RV. The patient remained stable and was discharged home 4 days after the procedure. The second patient presented with septic shock with evidence of lead associated endocarditis. Lead extraction was successful with no intraprocedural complications. However, she remained persistently septicemic with evidence of worsening end organ dysfunction. The family elected to withdraw life support measures on hospital day 4 due to her poor prognosis. The 6 minor complications included pocket hematomas, upper extremity swelling and need for blood transfusion. The Riata group had 5 complications including a major complication related to migration of a lead-associated thrombus to the main pulmonary artery. This patient remained hemodynamically stable and clot retrieval was performed the next day using a percutaneous snare technique. This patient was initially offered surgical extraction but a transvenous extraction was undertaken due to his unwillingness for an elective cardiac surgery. The 4 minor complications included pocket hematomas and need for blood transfusion. There was no statistically significant difference in median length of hospital stay or 30-day mortality (Fidelis® 2.1%, Riata 2.1%, P = 0.98).
Exploratory Analyses
Due to the higher prevalence of device infection in the Riata group, an analysis was performed excluding all patients with cardiac device-related infection. Results were similar to the primary analysis, with no statistically significant differences observed in procedure duration, procedural success, length of stay, complications, or 30-day mortality (Table 4).
Procedural characteristics and complications were compared among patients with Riata lead extractions by degree of conductor externalization. Twenty seven Riata patients had types 0 or 1 externalization and 20 had types 2 or 3. Patients with type 2 or 3 externalizations required the use of larger laser sheaths (14 Fr for type 0/1 vs. 16 Fr for type 2/3, P = 0.02). There were no statistically significant differences in number of sheaths used, procedural duration, procedural success, complications, length of hospital stay, or 30-day mortality (Table 5). Riata patients with type 2 or 3 externalization (n = 20) were compared with Riata and Fidelis® patients with no or minor (type 0 or 1) externalization (n = 172). Extractions with types 2 or 3 externalization required the use of larger laser sheaths (12 Fr for 0/1 vs 16 Fr for 2/3, P<0.001). There were no significant differences in number of sheaths used, procedural duration, procedural success, complications, length of stay, or 30-day mortality (Table 6).
Extractions of either lead type were compared based on the presence or absence of electrical lead failure (n = 122 vs. n = 70, respectively). Extractions of leads without electrical failure required larger laser sheaths (14 Fr vs 12 Fr, P<0.001), had a lower rate of procedural success (94.3% vs 99.2%, P = 0.04), a higher incidence of complications (11.4% vs 4.1%, P = 0.05), longer hospital stays (2 days vs 1 day, P = 0.001), and higher 30-day mortality (5.7% vs 0%, P = 0.008, Table 6). Leads without electrical failure were implanted for a significantly longer duration: 71% of leads without electrical failure had been implanted for more than 5 years, compared with 49.2% of leads with electrical failure (P = 0.001).