Abstract
In this retrospective study we compared different lead extraction techniques.
Between January 2009 and December 2012 we performed transvenous lead extraction procedures on 206 leads in 122 patients. Mean implant duration (MID) was 69.6 months (1–384 months). Leads with lead implant duration ≥ 12 months were assigned to groups according to the extraction technique: Group A: no extraction tool; Group B: laser approach; and Group C: mechanical approach. Overall clinical success was 93.3%. Group A showed a significantly lower MID [38.1 (19–122) months] compared with Groups B and C [83.1 (13–168) months; P < 0.0001 vs. 95.4 (12–384) months; P < 0.0001]. Mean implant duration between Groups B and C did not differ significantly (P = 0.28). Clinical and complete procedural success was 100% in Group A. Clinical success rate was higher in Group C than in Group B (97.0 vs. 76.9%, P = 0.018). Complete procedural success did not differ significantly between Groups B and C (88.9 vs. 76.9%; P = 0.132). In Groups B and C, absence of complete procedural success occurred in long implanted leads (MID 107.8 ± 36.4 and 137.6 ± 89.2 months). Relative costs per extracted lead were 49% higher in Group B than in Group C.
In case of long implanted leads a laser and a mechanical approach are comparable in complete procedural success and safety. Clinical success and cost effectiveness analysis favours the mechanical approach. Regardless of the extraction technique efficacy and safety optimization has to focus on long implanted leads.
With recent high-end mechanical extraction tools we showed that efficacy and safety results of a mechanical approach vs. a laser approach are comparable, whereas in the past usually higher success rates were reported in laser extraction procedures compared with mechanical approaches.
Cost effectiveness analysis favours a staged mechanical approach compared with a laser approach.
Introduction
During the past decades an extensive increase in the number of cardiac implantable electronic devices (CIED) is reported.1 Worldwide ∼500 000 CIEDs and 1 million leads are implanted annually.2,3 Concomitantly a disproportional increase in device infections is recognized.1 Deep pocket infections as well as lead endocarditis necessitate complete removal of the pacemaker or implantable cardioverter defibrillator (ICD) systems in order to allow the cure of such infections.4–6
Infections represent one of five indication groups for lead extractions according to the Expert Consensus Statement on transvenous lead extraction published by the Heart Rhythm Society in 2009.6 This consensus statement also defines chronic pain, thrombosis and venous stenosis, as well as functional and non-functional leads as indication classes for transvenous lead extractions. Compared with the NASPE recommendations for lead extraction published in 2000 the number of possible indications has hence clearly increased.7
Besides infections, lead failure is a commonly encountered problem in daily practice, which gains increasing importance in the light of lead extraction procedures.8
This explains why lead extractions have increased over the past decade and are likely to continue to increase. Therefore, techniques and tools of transvenous lead extractions have become more and more important. Several studies have shown high success rates and low complication rates of such procedures when standard extraction protocols are applied.3,9–19 Optimization of lead extraction techniques and tools remains an important issue. Part of this process is the comparison of available techniques and instruments.
The aim of this retrospective study was to analyse and compare the results of different extraction techniques and tools.
Methods
Patient population
Between January 2009 and December 2012 transvenous lead extraction procedures were performed on 122 patients. Mean patient age was 60.4 years (23–89 years), 87 were male and 35 female. 206 leads (122 pacemaker and 84 ICD leads) were planned for extraction due to different indications (Table 1) with an average of 1.7 leads per procedure. Mean implant duration (MID) of the leads was 69.6 months, ranging from 1 to 384 months. One hundred and forty-two were active fixation leads and 64 were passive fixation leads. Among the 84 ICD leads 75 showed a dual coil and 9 a single coil. Sixty-two leads were right atrial leads, 122 right ventricular leads, and 22 left ventricular leads.
Indications for lead extraction of all lead extraction procedures between 2009 and 2012 with regard to the different indication groups defined in the Expert Consensus of the Heart Rhythm Society
| Indication group | Number of leads | Percentage |
|---|---|---|
| Infection | 85 | 41.3 |
| Chronic pain | 0 | 0 |
| Thrombosis and venous stenosis | 8 | 3.9 |
| Non-functional leads | 105 | 50.9 |
| Functional leads | 8 | 3.9 |
| Indication group | Number of leads | Percentage |
|---|---|---|
| Infection | 85 | 41.3 |
| Chronic pain | 0 | 0 |
| Thrombosis and venous stenosis | 8 | 3.9 |
| Non-functional leads | 105 | 50.9 |
| Functional leads | 8 | 3.9 |
Indications for lead extraction of all lead extraction procedures between 2009 and 2012 with regard to the different indication groups defined in the Expert Consensus of the Heart Rhythm Society
| Indication group | Number of leads | Percentage |
|---|---|---|
| Infection | 85 | 41.3 |
| Chronic pain | 0 | 0 |
| Thrombosis and venous stenosis | 8 | 3.9 |
| Non-functional leads | 105 | 50.9 |
| Functional leads | 8 | 3.9 |
| Indication group | Number of leads | Percentage |
|---|---|---|
| Infection | 85 | 41.3 |
| Chronic pain | 0 | 0 |
| Thrombosis and venous stenosis | 8 | 3.9 |
| Non-functional leads | 105 | 50.9 |
| Functional leads | 8 | 3.9 |
The annual numbers of extraction showed a relevant increase in the years 2011 and 2012 compared with 2009 and 2010 (Figure 1).
Annual numbers of lead extractions between 2009 and 2012.
Retrospectively, leads with an implant duration of 12 months and more (179 leads, MID 79.6 months, range: 12–384 months) were assigned to three different groups according to the required extraction tools. Group A included all leads that were extracted by simple traction without the help of specific extraction instruments. In contrast, Group B included all leads that were extracted using a laser extraction approach, while leads in Group C were extracted with a mechanical extraction approach. Appropriate institutional review board approval was obtained. The results of the different groups were analysed and compared.
Extraction procedure
From January 2009 until December 2010 virtually all patients undergoing lead extractions had a laser extraction approach if leads could not be extracted by simple traction. A lead lock device (LLD, Spectranetics) was used for the application of traction and a laser sheath (SLS II, Spectranetics) for the application of counterpressure or countertraction.
With a change in the performing physician a staged extraction approach was introduced in December 2010 as part of a standard extraction protocol (Figure 2). This staged approach mainly uses mechanical extraction tools, but also allows the use of a laser sheath at the stage of powered extraction sheaths. For application of traction a locking stylet was used (Liberator, Cook Medical or LLD, Spectranetics). To apply counterpressure or countertraction either a simple extraction sheath (Byrd Dilator Sheath, Cook Medical) or in a next step a mechanical dilator sheath (Evolution, Cook Medical) or a laser sheath (SLS II, Spectranetics) could be used. Even though allowed by the extraction protocol, a crossover at the stage of the powered extraction sheaths from mechanical to laser or vice versa was never performed. In case of failed or impossible cranial approach, a femoral approach using a snare (Needle's Eye Snare, Cook Medical) was performed. Lead extraction procedures were performed in an operating room until April 2011, thereafter in a hybrid operating room in general anaesthesia by a cardiac surgeon with standby of extracorporeal circulation and a perfusionist. Patients were completely monitored including ECG, invasive blood pressure measurement, pulse oximetry, and transesophageal echocardiography.
Schematic illustration of the staged extraction approach used in our institution.
Definition of success, failure, and complications
Success was defined either as complete procedural success or as clinical success as defined by the expert consensus statement of the Heart Rhythm Society.6 Complete procedural success represented removal of all lead material confirmed by fluoroscopy with the absence of any permanently disabling complication or procedure-related death. Clinical success was defined as removal of all lead material or retention of a small portion of the lead that does not negatively affect the outcome goals of the procedure. Absence of clinical success was categorized as failure of the procedure.
Complications were described as minor or major according to the expert consensus on transvenous lead extractions of the Heart Rhythm Society.6 Complication rates were calculated as the number of complications in relation to the overall number of patients treated.
Statistics
Data were analysed retrospectively using the SPSS software Version 20 (IBM Corporation). Categorical variables were presented as numbers and percentages. Continuous variables are presented as mean ± standard deviation or as mean with range. To address clustering of leads within patients, linear and logistic regressions with robust standard error and patient-ID as cluster were performed to analyse the differences between groups in complete removal, implant duration, and percentage of ICD leads. Linear and logistic regression with robust standard errors were used to analyse the relations between the groups—Stata 11 (StataCorp) was used for this regression. A P value (P) <0.05 was considered statistically significant.
Results
Among all leads with an implant duration of 12 months and more, clinical success rate was 93.7%, with complete procedural success in 88.8%. Operative mortality was zero.
Two major complications occurred during all extraction procedures (1.6%). One patient suffered from exit block due a dislocated right ventricular lead by a femoral attempt of extraction with a snare retrieval device requiring a short period of chest compressions until re-establishment of right ventricular pacing. The patient recovered without any residuals. Another patient had ventricular rupture and required immediate thoracotomy. The patient survived.
Minor complications were seen in seven cases (5.7%): six patients with pocket hematomas which required surgical intervention and one patient with a pneumothorax which required chest tube placement for drainage.
Leads in Group A showed a significantly lower mean time since implantation (38.1 months, range: 19–122 months) compared with Group B (83.1 months, range: 13–168 months; P < 0.0001) and Group C (95.4 months, range: 12–384 months; P < 0.0001). There was no difference in MID between Groups B and C (P = 0.28) (Figure 3).
Lead extraction procedures can be very challenging due to severe adhesions.
The number of ICD leads was significantly higher in Group B (69.2%) than in Group A (38.0%; P = 0.007) and Group C (30.6%; P = 0.001).
The rate of clinical and complete procedural success was 100% in Group A. Complete procedural success was numerically better in Group C than in Group B without statistical significance (P = 0.132).
Clinical success rate was higher in Group C than in Group B (97.0 vs. 76.9%, P = 0.018) (Table 2).
Results of leads with an implant duration of 12 months and more with regard to the different groups (group A = no specific extraction tools, Group B = laser extraction approach, Group C = mechanical extraction approach; n.s. = not significant)
| Group A (n = 41) | Group B (n = 39) | Group C (n = 99) | P value | |
|---|---|---|---|---|
| Mean implant duration (months) | 38.1 (19–122) | 83.1 (13–168) | 95.4 (12–384) | A vs. B: P < 0.0001 A vs. C: P < 0.0001 B vs. C: n.s. |
| Ratio of ICD leads | 26.8% | 69.2% | 37.4% | A vs. B: P < 0.0001 A vs. C: n.s. B vs. C: P = 0.001 |
| Complete procedural success | 100% (41) | 76.9% (30) | 88.9% (88) | A vs. B: P < 0.0001 A vs. C: P = 0.005 B vs. C: n.s. |
| Clinical success | 100% (41) | 76.9% (30) | 97.0% (96) | A vs. B: P = 0.001 A vs. C: n.s. B vs. C: P = 0.018 |
| Operative Mortality | 0 | 0 | 0 | n.s. |
| Minor complications | 2 | 2 | 3 | n.s. |
| Major complications | 0 | 0 | 2 | n.s. |
| Group A (n = 41) | Group B (n = 39) | Group C (n = 99) | P value | |
|---|---|---|---|---|
| Mean implant duration (months) | 38.1 (19–122) | 83.1 (13–168) | 95.4 (12–384) | A vs. B: P < 0.0001 A vs. C: P < 0.0001 B vs. C: n.s. |
| Ratio of ICD leads | 26.8% | 69.2% | 37.4% | A vs. B: P < 0.0001 A vs. C: n.s. B vs. C: P = 0.001 |
| Complete procedural success | 100% (41) | 76.9% (30) | 88.9% (88) | A vs. B: P < 0.0001 A vs. C: P = 0.005 B vs. C: n.s. |
| Clinical success | 100% (41) | 76.9% (30) | 97.0% (96) | A vs. B: P = 0.001 A vs. C: n.s. B vs. C: P = 0.018 |
| Operative Mortality | 0 | 0 | 0 | n.s. |
| Minor complications | 2 | 2 | 3 | n.s. |
| Major complications | 0 | 0 | 2 | n.s. |
Results of leads with an implant duration of 12 months and more with regard to the different groups (group A = no specific extraction tools, Group B = laser extraction approach, Group C = mechanical extraction approach; n.s. = not significant)
| Group A (n = 41) | Group B (n = 39) | Group C (n = 99) | P value | |
|---|---|---|---|---|
| Mean implant duration (months) | 38.1 (19–122) | 83.1 (13–168) | 95.4 (12–384) | A vs. B: P < 0.0001 A vs. C: P < 0.0001 B vs. C: n.s. |
| Ratio of ICD leads | 26.8% | 69.2% | 37.4% | A vs. B: P < 0.0001 A vs. C: n.s. B vs. C: P = 0.001 |
| Complete procedural success | 100% (41) | 76.9% (30) | 88.9% (88) | A vs. B: P < 0.0001 A vs. C: P = 0.005 B vs. C: n.s. |
| Clinical success | 100% (41) | 76.9% (30) | 97.0% (96) | A vs. B: P = 0.001 A vs. C: n.s. B vs. C: P = 0.018 |
| Operative Mortality | 0 | 0 | 0 | n.s. |
| Minor complications | 2 | 2 | 3 | n.s. |
| Major complications | 0 | 0 | 2 | n.s. |
| Group A (n = 41) | Group B (n = 39) | Group C (n = 99) | P value | |
|---|---|---|---|---|
| Mean implant duration (months) | 38.1 (19–122) | 83.1 (13–168) | 95.4 (12–384) | A vs. B: P < 0.0001 A vs. C: P < 0.0001 B vs. C: n.s. |
| Ratio of ICD leads | 26.8% | 69.2% | 37.4% | A vs. B: P < 0.0001 A vs. C: n.s. B vs. C: P = 0.001 |
| Complete procedural success | 100% (41) | 76.9% (30) | 88.9% (88) | A vs. B: P < 0.0001 A vs. C: P = 0.005 B vs. C: n.s. |
| Clinical success | 100% (41) | 76.9% (30) | 97.0% (96) | A vs. B: P = 0.001 A vs. C: n.s. B vs. C: P = 0.018 |
| Operative Mortality | 0 | 0 | 0 | n.s. |
| Minor complications | 2 | 2 | 3 | n.s. |
| Major complications | 0 | 0 | 2 | n.s. |
With regard to presence or absence of complete procedural success MIDs were compared in the laser and the mechanical extraction approach group. In both groups (B and C) MID was higher in the group without than in the group with complete procedural success. Mean implant duration in the extractions without complete procedural success were 9 years and more regardless of the extraction approach (laser or mechanical) (Table 3).
Comparison of mean implant duration with regard to presence or absence of complete procedural success in lead extractions with laser or mechanical approach
| Group B (laser approach) (n = 39) | Group C (mechanical approach) (n = 99) | |
|---|---|---|
| Mean implant duration (months) of leads with complete procedural success | 75.7 ± 52.0 (n = 30) | 90.0 ± 50.3 (n = 88) |
| Mean implant duration (months) of leads with absence of complete procedural success | 107.8 ± 36.4 (n = 9) | 137.6 ± 89.2 (n = 11) |
| Group B (laser approach) (n = 39) | Group C (mechanical approach) (n = 99) | |
|---|---|---|
| Mean implant duration (months) of leads with complete procedural success | 75.7 ± 52.0 (n = 30) | 90.0 ± 50.3 (n = 88) |
| Mean implant duration (months) of leads with absence of complete procedural success | 107.8 ± 36.4 (n = 9) | 137.6 ± 89.2 (n = 11) |
Comparison of mean implant duration with regard to presence or absence of complete procedural success in lead extractions with laser or mechanical approach
| Group B (laser approach) (n = 39) | Group C (mechanical approach) (n = 99) | |
|---|---|---|
| Mean implant duration (months) of leads with complete procedural success | 75.7 ± 52.0 (n = 30) | 90.0 ± 50.3 (n = 88) |
| Mean implant duration (months) of leads with absence of complete procedural success | 107.8 ± 36.4 (n = 9) | 137.6 ± 89.2 (n = 11) |
| Group B (laser approach) (n = 39) | Group C (mechanical approach) (n = 99) | |
|---|---|---|
| Mean implant duration (months) of leads with complete procedural success | 75.7 ± 52.0 (n = 30) | 90.0 ± 50.3 (n = 88) |
| Mean implant duration (months) of leads with absence of complete procedural success | 107.8 ± 36.4 (n = 9) | 137.6 ± 89.2 (n = 11) |
Major and minor complications were low in all groups and showed no statistically significant differences.
With regard to the costs per extracted lead relative costs were calculated. With regard to the mechanical approach costs were calculated by summing up the costs of the extraction tools and dividing them by the number of extracted leads. In this group there was no need to add maintenance costs to the calculation, since only single-use products were used. In the laser extraction group in addition to the costs of the used extraction tools also the yearly maintenance costs of the laser system itself were included. The cost per extracted lead using the mechanical approach was defined as 100. In relation to this base rate the relative cost of the laser extraction approach was calculated as 149.
Discussion
Due to the increase in CIED implantations during the past decades, the actual number of lead extraction procedures worldwide, the durability of leads and the life expectation of patients the necessity for lead extraction procedures will continue to increase.1,8,20,21
Nowadays the reported success rates of transvenous lead extraction procedures are high with low complication rates and minimal mortality.3,9–19 These impressive results should not mislead physicians over the fact that if complications occur during such procedures they are often disastrous and put patients at risk of their life. Even with modern techniques and tools, there are still a certain number of extractions which are very challenging (Figure 3) or end with failure. Therefore, it is of utmost importance to continue to optimize lead extraction techniques and tools.
Lead extraction procedures may be very cost-intensive especially in very challenging cases where several extraction tools or expensive high-end tools are needed to achieve success. In times of economically struggling health-care systems worldwide, the cost factor becomes more and more important in the treatment of patients. In summary, optimization efforts of lead extraction procedures have to involve safety and efficacy aspects as well as economic aspects.
As a first step in this optimization process actual techniques should be evaluated and compared.
Our results and complication rates compare favourably to the results presented in the literature over the past years.3,9–17 Some studies reported higher success rates than we were able to achieve.12,18,19 In those studies, however, MID was shorter than in our groups with the laser or mechanical approach.
With regard to the different indications most lead extraction procedures were related to infections or non-functional leads approximately at a 1 : 1 ratio accounting for a little more than 90% of the leads (Table 1). Infection as a large indication group is consistent with the literature.12,16–19 With regard to non-functional leads we follow a proactive lead extraction philosophy based on an individual risk–benefit analysis for every single patient. Lead failure is a scenario which is increasingly encountered in clinical practice. Kleemann et al.8 were able to show that the annual rate of ICD lead failure increased with time and reached 20% after 10 years of implant time. Taking this failure rate into account in combination with the annual numbers of lead implantations, it becomes evident that this indication group will increasingly become important. Abandoning non-functional leads carries potential future disadvantages for the patient especially due to a higher procedural risk and a lower success rate for necessary extraction procedures in the future. This is caused by more severe adhesions due to longer implant times and possibly by more implanted leads with lead–lead interactions.
During our observational interval, our lead extraction protocol changed due to a change in the performing physician from a laser approach to a staged, predominantly mechanical approach (Figure 1), which gave us the possibility to retrospectively assign the leads to different groups with regard to their extraction technique in order to analyse differences.
Leads that were extracted by simple traction without the help of any specific extraction tool showed, as expected, a significantly shorter implant duration. Furthermore, this group had a 100% success rate, certainly due to the short implant duration and not in consequence of the technique. Therefore, this group of patients certainly is not the focus for safety, efficacy or economical optimization efforts.
The results for complete procedural success were comparable between the laser and the mechanical extraction approach. Numerically complete procedural success was higher in the mechanical approach without statistically significant differences. The clinical success rate showed statistically significant better results in the mechanical than in the laser extraction group. This finding is not entirely consistent with the published literature which usually shows higher rates of complete removal in laser extraction procedures than in mechanical approaches.9,12,17–19 However, there are only few direct comparisons between laser and mechanical approaches not including recent high-end mechanical extraction devices like the Evolution or the Evolution Shortie (Cook Medical) mechanical dilator sheaths. In the context of the trend for better results in the mechanical approach group it is important to mention the fact that in the laser group the percentage of ICD leads was significantly higher.
With regard to major or minor complications there were no statistically significant differences between the laser and mechanical extraction approach.
Since success rates in both the laser and the mechanical extraction group are <100% there certainly is potential for efficacy optimization. This in mind we compared the MIDs of successful with failed procedures in both groups. Indeed, MID was higher in the failed in comparison with the successful lead extractions regardless of the approach. In both groups MID of leads with extraction failure was greater than or approaching ten years. This result is consistent with the results of the LEXICON study, where a progressive increase in extraction failure was seen in leads with long implant durations.19 Therefore, efforts of efficacy optimization of techniques and tools should mainly focus on extraction procedures of leads, which have been implanted for 10 years and more.
When comparing relative costs per extracted lead, 49% higher costs were identified in the laser group. In our opinion this advantage in relative costs in our study is mainly due to the staged approach in our mechanical extraction group which offers the surgeon the possibility to use cheaper extraction tools in the first line and only escalating to expensive high-end tools in case of severe adhesions making the extraction impossible with simple tools. It is also important to mention that regarding mechanically or laser-powered extraction sheaths, the laser approach has the disadvantage of a double investment into the sheath itself and an excimer laser system, whereas the mechanical dilator sheath does not need any additional device.
The limitations of this study are its retrospective study design and the fact that it is a single-centre experience. A limitation with regard to the comparison of the laser and the mechanical approach is the fact that during the time of the laser approach (2009 and 2010) the annual volume of lead extractions was lower than during the time period of the mechanical approach (2011 and 2012). To further evaluate and optimize lead extraction techniques and tools more randomized prospective multi-centre trials are needed.
Conclusions
In case of leads with a long implant duration, the rates for complete procedural success are comparable between the laser and the mechanical extraction approach. Clinical success is better with the mechanical approach. Complication rates do not differ significantly between the two approaches. Cost-effectiveness analysis favours the mechanical approach. Efficacy and safety optimization efforts for techniques and tools have to focus on the extraction of long implanted leads. Efforts on economic optimization have to concentrate on staged approaches.
Conflict of interest: C.T.S. and D.H. receive workshop honoraria from Cook Medical Europe Ltd. The other authors declare no conflict of interest.
Acknowledgements
We like to express our gratitude to Prof Dr Burkhardt Seifert from the Biostatistics Unit, Institute of Social and Preventive Medicine, University of Zurich, Switzerland for performing the statistical analysis of this study.
