Successful avoidance of superior vena cava injury during transvenous lead extraction using a tandem femoral-superior approach

Abstract

BACKGROUND

Transvenous pacemaker and defibrillator lead extraction is a higher risk procedure with variation in preferred technique. A frequently fatal complication of this procedure is perforation of the superior vena cava. We have developed a tandem femoral-superior technique that incorporates snaring of targeted leads from a femoral approach combined with use of a rotational cutting sheath advanced over the lead from the subclavian vein.

OBJECTIVE

We sought to evaluate the safety and efficacy of a tandem femoral-superior approach to lead extraction.

METHODS

Consecutive patients undergoing transvenous extraction of at least 1 pacemaker or defibrillator lead with implant duration ≥1 year in which a tandem femoral-superior technique was used as the initial extraction strategy were included. The registry spanned 2010–2018 and consisted of procedures performed by a single primary operator.

RESULTS

A total of 131 patients were included. A total of 267 leads with a mean implant duration of 9.8 years, including 90 defibrillator leads (33.7%), were targeted for extraction. No superior vena cava perforation or other vascular damage occurred. Clinical procedural success was achieved in 96.2% of cases. There were 5 major complications (3.8% of patients), with 3 being pericardial effusion requiring intervention. There were no deaths.

CONCLUSION

A tandem femoral-superior approach to lead extraction effectively eliminated superior vena cava injury. This is a safe and effective technique for transvenous lead extraction.

Introduction

An ongoing challenge in the management of patients with pacemakers and defibrillators is the increasing need to remove intravascular leads. This is most commonly performed percutaneously by transvenous lead extraction.^1–2 Intravascular and intracardiac fibrosis can make this a potentially dangerous procedure. A variety of tools and techniques have been developed to aid in removal of these leads including telescoping sheaths, rotational sheaths, laser sheaths, locking stylets, and snares deployed from the femoral and jugular veins.^3–8 Rotational or laser sheaths advanced over leads from the vein entry site have become the dominant tool for lead extraction in the United States. Femoral snares are most commonly used only as a backup technique when other approaches have failed.^5–9–10

A particularly feared complication of transvenous lead extraction is perforation of vascular structures, most notably of the superior vena cava, which may result in uncompressible hemorrhage into the right thorax and in 1 study carried a mortality risk of 44%.¹¹ Remarkably, perforation of the superior vena cava was implicated in 70% of all deaths associated with lead extraction procedures recently reported to the US Manufacturer and User Facility Device Experience database.⁹ This has led to efforts to mitigate the risk of vascular tear including routine use of bridging balloons¹² or intraprocedural thoracoscopy.¹³

The University of Utah Hospital is a high-volume center for lead extraction taking referrals from a wide geographic region for higher risk procedures. This referral pattern has led to an extraction population weighted toward patients with leads of longer implant duration and with other high risk features. In contradistinction to many extraction programs, our program has used a femoral snare as a component of the initial approach for nearly all extraction procedures. We report here the safety and efficacy of a tandem technique for extraction using a combination of a femoral snare and a superior sheath. We hypothesized that this technique would lead to fewer vascular injuries, including superior vena cava injury.

Methods

Study design

We performed a retrospective analysis of all patients who underwent transvenous lead extraction using a tandem femoral-superior technique as the initial approach at the University of Utah Hospital between 2010 and 2018. All patients had at least 1 lead that had been implanted for a year or longer. As a single primary operator (R.A.F.) participated in nearly all lead extraction procedures during that time period, we further excluded any cases where he was not the primary operator. All analyses of the data and reporting of anonymous aggregate data were approved by the University of Utah Institutional Review Board, with a common rule exemption to informed consent of participants. The research reported in this article adhered to the Helsinki Declaration as revised in 2013.

Data collection

Patients were identified using a combination of procedure case log review and use of the University of Utah’s enterprise data warehouse. Patients were initially identified by screening for Current Procedure Terminology codes 33234, 33235, 33244, and 33272 between 2010 and 2018. Additionally, starting in 2014 when the University of Utah Hospital implemented the Epic electronic medical record system (Epic Systems, Verona, WI), patients were further identified by screening Epic’s procedure supply for use of femoral extraction tools.

For all potentially qualifying cases, procedure notes were reviewed in detail to confirm the technique used as well as for intraprocedural complications. When necessary, radiographs were reviewed to clarify the size of retained lead fragments. Postprocedure echocardiograms were reviewed to clarify the extent of any change in tricuspid regurgitation resulting from the extraction procedure. Progress notes and discharge summaries were reviewed for complications identified during the index hospitalization.

Statistical methods

Patient baseline characteristics, lead characteristics, and outcomes are presented using standard summary statistics, including frequencies, percentages, and means. Measures of variation are presented as mean ± SD. Differences between patient groups for categorical variables were evaluated using the χ² test or the Fisher exact test, as appropriate, and continuous variables were evaluated using 2-group Student t tests. Statistical analyses were performed using Stata 16 (StataCorp LLC, College Station, TX).

Extraction technique

Procedures were performed in a hybrid operating room or, for some cases felt to be lower risk, in the electrophysiology laboratory. A preoperative chest radiograph and transthoracic echocardiogram were the only imaging tests uniformly obtained preprocedure, although patients with known or suspected device infection also had a preprocedure transesophageal echocardiogram. For extraction, general anesthesia along with transesophageal echocardiogram monitoring by a qualified cardiac anesthesiologist was uniformly performed. All patients had femoral arterial cannulae for blood pressure monitoring, femoral venous cannulae for fluid administration, and temporary pacing as needed. Appropriate blood products were in the room at the start of each case. Anticoagulation with warfarin was interrupted or reversed (goal INR ≤1.3), and direct thrombin inhibitors and P2Y12 inhibitors were discontinued for all procedures. Surgical and interventional cardiology backup was on standby for every procedure but not physically in the procedure room until needed. Superior vena cava occlusion balloons were not used.

The extraction procedure commenced with typical lead preparation from the pectoral implant site. The device pocket was opened, all leads freed to their suture sleeves, and the suture sleeves mobilized for leads targeted for extraction. Standard stylets were placed down all leads. For targeted leads with an extendable-retractable fixation helix, an attempt was made to retract the helix by rotating the lead’s proximal pin.

A femoral workstation (Cook Medical LLC, Bloomington, IN) was placed in the right femoral vein using ultrasound guidance. Rarely, the workstation was placed in the left femoral vein or the right internal jugular vein. The primary snaring device used was the Needle’s Eye Snare (Cook Medical LLC, Bloomington, IN), which was advanced through the workstation into the right atrium or superior vena cava where the target lead was snared but not tightly cinched so as not to compromise the lumen. Before the snare was cinched, a locking stylet was advanced down the lead as distally as possible, and then the lead was tightly cinched. In most cases the locking stylet was advanced well beyond the point where the lead was snared and thus the cinched snare provided an additional point of stability for the locking stylet. The lead and locking stylet were also secured close to the vein entry site using a suture or a compression coil (One-Tie, Cook Medical LLC). For any active fixation lead whose helix could not be retracted with rotation of the proximal pin, an attempt was made to unscrew the lead from the endocardium by rotating the snare counterclockwise. Traction was exerted on the lead using the snare from below until there was straightening of the lead in the superior vena cava and right atrium (Figure 1). A rotating dissection sheath system (Evolution, Evolution RL, Evolution Shortie RL, and/or SteadySheath, Cook Medical LLC) was advanced over the locking stylet and the lead into the vasculature. Other sheath systems, such as simple telescoping sheaths or laser sheaths, were not used. Once the superior vena cava was reached, typically the outer nonrotating sheath was then advanced in a piston motion over the lead to avoid unprotected use of the rotational cutting blade in this area, followed by the rotational sheath within the outer sheath. When the rotational sheath reached the point where the lead was snared from below, the snare was opened to allow the sheath to pass. The sheath was further advanced toward the distal end of the lead until the lead was free of its endocardial attachment. The lead was then removed either through the sheath from above or using the snare from below.

Figure 1.

Tandem femoral-superior extraction technique applied to a right ventricular lead. The thick solid arrow shows the natural direction of the rotational sheath (A) as it is advanced toward the lateral wall of the superior vena cava. Traction is applied on the lead superiorly (short-dash arrow) via the locking stylet (D) through the rotational sheath and simultaneously with the femoral snare (B) in an inferior direction (long-dash arrow). The result of simultaneous traction from above and below is to pull the lead (C) away from the lateral wall of the superior vena cava (short thin arrow). Traction also results in the lead serving as a more stable rail for advancement of the rotational sheath through adhesions and prevents superior traction from being transmitted to the point of myocardial attachment (E).

No active venous closure device was used after removal of the femoral snare system, but a figure-of-8 stitch was commonly used in the skin and subcutaneous tissues. Postoperatively patients had a 6-hour period of bed rest.

Study end points

Outcome definitions were in accordance with the recommendations of the European Heart Rhythm Association.⁷ Complete extraction success was defined on a per-lead basis and required no retained lead material visible radiographically. Clinical success was defined on a per-patient basis and required no more than 4 cm of retained lead material visible radiographically, achievement of all clinical objectives, and no major complications.

Classification of major and minor complications was consistent with those recommended by the Heart Rhythm Society⁶ and the European Heart Rhythm Association.⁷ Screened major complications included pericardial effusion requiring drainage; stroke; vascular perforation, avulsion, or tear; and death. Screened minor complications included new pericardial effusion not requiring intervention, hematoma requiring postprocedure evacuation, vascular repair at venous access site, migrated segment of the lead, increase in the degree of tricuspid regurgitation, and pneumothorax requiring a chest tube.

Results

Patient and lead characteristics

Between 2010 and 2018, a total of 131 patients underwent transvenous lead extraction using the tandem femoral-superior technique. Of these, 114 cases (87%) were performed in a hybrid operating room and 17 cases (13%) in the electrophysiology laboratory. The mean patient age was 62.2 ± 16.2 years, and 39 (29.8%) were female. The additional baseline characteristics of the patients are summarized in Table 1.

Table 1.

Characteristics of 131 patients

Characteristic	Value
Age (y)	62.2 ± 16.2
Male sex	92 (70.2)
BMI
Mean ± SD (kg/m2)	28.6 ± 6.4
<20 kg/m2	5 (4.0)
Ejection fraction (%)	47.7 ± 14.0
Primary cardiac diagnosis
Ischemic heart disease	37 (28.2)
Dilated cardiomyopathy	28 (21.4)
Primary electrical disease	43 (32.8)
Other	23 (17.6)
Comorbidities
Hypertension	68 (51.9)
Diabetes	32 (24.4)
Chronic kidney disease	21 (16.0)
Obstructive pulmonary disease	21 (16.0)
End-stage renal disease	2 (1.5)
Prior sternotomy	23 (17.6)
Indication for extraction
Infection	51 (38.9)
Systemic infection	20 (15.3)
Pocket infection only	31 (23.7)
Noninfectious	80 (61.1)
Device type
Pacemaker	55 (42.0)
ICD	32 (24.4)
CRT-D	44 (33.6)
Targeted leads per patient	2.0 ± 0.9

Values are presented as mean ± SD or n (%).

BMI = body mass index; CRT-D = cardiac resynchronization therapy – defibrillator; ICD = implantable cardiac defibrillator.

A total of 267 leads were targeted for extraction. Of these, 90 (33.7%) were implantable cardiac defibrillator leads; 219 (82.2%) were active fixation. The mean implant duration was 9.8 years with the longest implant duration of 35.7 years. The additional characteristics of leads are summarized in Table 2.

Table 2.

Characteristics of 267 targeted leads

Characteristic	Value
Pacemaker leads	177 (66.3)
Defibrillator leads	90 (33.7)
Single coil	25 (14.1)
Dual coil	65 (36.7)
Lead location
Right ventricle	142 (53.2)
Right atrium	97 (36.3)
Coronary sinus	28 (10.5)
Fixation type
Active	219 (82.0)
Passive	48 (18.0)
Implant duration
Mean ± SD (y)	9.8 ± 5.5
≤2 y	18 (6.7)
>2–≤5 y	31 (11.6)
>5–≤10 y	104 (39.0)
>10–≤20 y	98 (36.7)
>20 y	16 (6.0)

Values are presented as mean ± SD or n (%).

Efficacy and safety outcomes

Clinical procedural success was achieved in 126 of 131 patients (96.2%). Complete extraction success was achieved in 246 of 267 targeted leads (92.1%).

No patient in our series experienced vascular injury, and in particular there was no tear, perforation, or other apparent damage to the superior vena cava. Four patients experienced a total of 5 major complications (Table 3). Three patients developed pericardial effusion requiring drainage. Two strokes occurred during or within 24 hours after the procedure. There were no procedural deaths.

Table 3.

Major and minor complications

Major complications	4/131 (3.1)
Procedure-related death	0 (0)
Pericardial effusion requiring percutaneous or surgical drainage	3 (2.3)
Procedure-related stroke	2 (1.5)
Vascular perforation, avulsion, or tear	0 (0)
Minor complications	11/131 (8.4)
Pocket hematoma requiring evacuation	3 (2.3)
Intravascular repair at the entry site	0 (0)
Migrated segment of the lead	1 (0.8)
Increase in the degree of tricuspid regurgitation	6 (4.6)
Pneumothorax requiring a chest tube	1 (0.8)

Values are presented as n/total patients (%) or n (%).

Complications are reported on a per-patient basis. One patient had 2 major complications.

Minor complications occurred in 11 of 131 patients (8.4%) (Table 3). More than half of these minor complications were an increase in the grade of tricuspid regurgitation without the need for immediate intervention. Most frequently, this was a finding reported on the intraoperative transesophageal echocardiogram without clinical correlation. However, 1 patient required tricuspid valve replacement 5 years after his extraction procedure for right heart failure that could not be adequately managed medically.

Discussion

One of the most catastrophic complications of transvenous pacemaker or defibrillator lead extraction is superior vena cava perforation from tear, burn, or other injury and resulting in incompressible bleeding into the right thorax. A recent analysis of deaths reported to the US Manufacturer and User Facility Device Experience database showed remarkably that 126 of 180 deaths (70%) associated with lead extraction procedures over a 6-year period involved injury to the superior vena cava.⁹ Our experience suggests that this risk can be reduced through use of a tandem femoral-superior approach to transvenous lead extraction.

Damage to the superior vena cava is thought to occur when the extraction sheath, advanced over the lead from the lead entry site, is inadvertently directed against the wall of the superior vena cava during attempts to turn it downward within the lumen of the superior vena cava. When a femoral snare is used along with a superior sheath in the tandem technique, traction from below has been shown by intracardiac echocardiography to pull the lead away from the lateral wall of the superior vena cava during passage of the superior sheath.¹⁴ This tandem technique is illustrated and detailed further in Figure 1. The robust rail provided by femoral traction prevents sudden misdirection of the superior sheath over the lead. Additionally, snaring the lead from below prevents the tension being exerted on the lead from above from being transmitted to the heart and thus obviates the hypotension and arrhythmias that otherwise frequently result. The rail also allows for more effective use of the outer nonrotating sheath (paired with the rotational sheath) to be advanced over the lead from above, especially in the area of the superior vena cava where we sought to minimize use of the cutting head of the rotational sheath. Consistent with these strategies is the observation of no instances of superior vena cava perforation in our series. Our data of 131 patients extracted using the tandem technique is the largest reported series of this technique to date.

Our tandem extraction technique evolved from an earlier extensive reliance on a femoral-only technique as the initial approach to extraction. Traction on a lead adhered to the superior vena cava with a snare from below has been shown by mathematical modeling to result on less stress applied to the wall of the superior vena cava than when traction is exerted from above.¹⁵ We still target some leads using a femoral-only approach. These include some lead models (eg, Fine-line, Boston Scientific Corp., Marlborough, MA; SelectSecure, Medtronic PLC, Dublin, Ireland) that are more amenable to extraction by a femoral-only approach on the basis of their design and our prior experience. The femoral-only approach is also necessarily used to extract leads that had been previously cut and are not accessible from the pocket.

Our tandem femoral-superior extraction techniques yielded safety and efficacy comparable to other contemporary reports of lead extraction series that rely more on the superior approach. The largest of these, the multicenter European Lead Extraction ConTRolled study, reported a clinical success rate of 96.7% and a procedure-related major complication rate of 1.7%, including a vascular avulsion event rate of 0.6%.⁵ The multicenter Lead Extraction in the Contemporary Setting study reported a clinical success rate of 97.7% and a procedure-related major complication rate of 1.4%, including an incidence of vascular tear of 0.4%.¹¹ More recent studies that exclusively used rotational sheaths showed similar success and complication rates but no injuries to the extrapericardial superior vena cava.^16–18 Our results indicate that similar outcomes can be achieved with a tandem femoral-superior strategy. Notably, the mean implant duration of leads targeted in the present study was 9.8 years, longer than other published series.^{5–11–16–18}

Our series was restricted to the use of a single type of the superior sheath, that is, a rotational sheath. In particular, the laser-powered sheath (GlideLight, Philips, Amsterdam, The Netherlands) was not used. Compared with the multicenter Lead Extraction in the Contemporary Setting study of the laser-powered sheath,¹¹ our series had higher incidences of pericardial effusion requiring drainage (2.3% vs 0.6%) and overall major complications (3.1% vs 1.4%). It is possible that with use of other superior sheath systems, the tandem femoral-superior technique would have a lower incidence of pericardial effusion and other major complications. Of note, there has been a prior report of 15 patients treated with a femoral-superior extraction approach using a laser-powered sheath successfully with no complications.¹⁴

Limitations

Our tandem femoral-superior strategy was not concurrently compared with the superior-only technique. However, the baseline patient and lead characteristics of our series are comparable to other contemporary lead extraction series that almost exclusively used superior-only technique, with the exception of longer implant durations in the present study than in other studies.

Since we uniformly used a rotational sheath and did not include use of other superior sheath systems such as simple telescoping sheaths or laser sheaths, our results may not be generalizable when other sheath systems are used, although as noted above, successful use of the femoral-superior technique with a laser sheath system has been reported.¹⁴ Finally, our series was confined to the cases performed by a single experienced primary operator and may not be generalizable to other operators.

We did not have data on procedure time for our tandem femoral-superior extraction strategy. Additional time is required to snare the lead from below, but possibly this is offset by the rapidity of advancing the sheath from above over the lead once snared. Overall, the effect of the tandem approach use on procedure time is uncertain.

Conclusion

The tandem femoral-superior approach to lead extraction effectively eliminated superior vena cava injury in a large series of lead extraction procedures. This is a safe and effective technique for transvenous lead extraction, with success and complication rates similar to those in contemporary reports of lead extraction, which rely mainly on a superior approach. Our experience supports the use of the tandem femoral-superior strategy as an alternate primary approach to lead extraction.