- Email: [email protected]
Outcomes of patients requiring emergent surgical or endovascular intervention for catastrophic complications during transvenous lead extraction
Outcomes of patients requiring emergent surgical or endovascular intervention for catastrophic complications during transvenous lead extraction Michael P. Brunner, MD,* Edmond M. Cronin, MB, MRCPI,* Oussama Wazni, MD,* Bryan Baranowski, MD,* Walid I. Saliba, MD, FHRS,* Joseph F. Sabik, MD,† Bruce D. Lindsay, MD, FHRS,* Bruce L. Wilkoff, MD, FHRS,* Khaldoun G. Tarakji, MD, MPH, FHRS* From the Departments of *Cardiovascular Medicine and †Cardiothoracic Surgery, Cleveland Clinic, Cleveland, Ohio. BACKGROUND The outcomes of patients requiring emergent surgical or endovascular intervention during transvenous lead extraction (TLE) have not been well characterized. OBJECTIVES To evaluate the incidence of catastrophic complications requiring emergent surgical or endovascular intervention during TLE, to describe the injuries, and to review patient management and outcomes. METHODS Consecutive patients undergoing TLE of pacemaker and implantable cardioverter-defibrillator (ICD) leads at the Cleveland Clinic between August 1996 and September 2012 were included in the analysis. RESULTS A total of 5973 (4436 [74.3%] pacemaker and 1537 [25.7%] ICD) leads were extracted during 3258 TLE procedures (median [25th, 75th percentile] patient age 67.0 [55.0, 76.1] years; 69.2% men). The median (25th, 75th percentile) lead implant duration was 4.9 (2.4, 8.4) years, and 2.0 (1.0, 2.0) leads were extracted per procedure. Powered sheaths were used in 2369 (72.7%) procedures. Twenty-five (0.8%) patients experienced catastrophic complications requiring emergent surgical or endovascular intervention. Twenty patients (0.6%) required either sternotomy
Introduction Cardiovascular implantable electronic device utilization has grown substantially in response to broadening evidencebased indications.1–3 In tandem, there has been a growing need for system revisions and lead extraction, driven by the Dr Saliba has received speaker’s honoraria from Boston Scientific, Medtronic, and St Jude Medical. Dr Lindsay is a member of the CardioInsight Technologies Physician Advisory Board. Dr Wilkoff has received speaker’s honoraria from Boston Scientific and St Jude Medical; he is also a member of the St Jude Medical Physician Advisory Board and Spectranetics Physician Advisory Board. Dr Tarakji has received speaker’s honoraria from HMP Communications. All authors with disclosures reported compensation less then $10,000 per year. Address reprint requests and correspondence: Dr Khaldoun G. Tarakji, Section of Cardiac Pacing and Electrophysiology, Heart and Vascular Institute, Cleveland Clinic, 9500 Euclid Avenue J2-2, Cleveland, OH 44195. E-mail address: [email protected].
1547-5271/$-see front matter B 2014 Heart Rhythm Society. All rights reserved.
(n ¼ 18) or thoracotomy (n ¼ 2) for superior vena cava laceration (n ¼ 15) and right atrial (n ¼ 2) or ventricular (n ¼ 3) perforation. Two patients required vascular repair at the procedural access site for either subclavian vein or artery laceration. Three patients were managed with an endovascular approach for superior vena cava laceration, left axillary artery laceration, and brachiocephalic vein and artery fistula. In-hospital mortality was 36.0% (6 procedural/operative deaths and 3 deaths during the same hospitalization). CONCLUSIONS Major vascular injury or cardiac perforation requiring emergent surgical or endovascular intervention during TLE is uncommon but carries significant in-hospital mortality. Despite high mortality, nearly two-thirds of these patients were rescued with immediate response and surgical or endovascular intervention. KEYWORDS Implantable cardioverter-defibrillator; Pacemaker; Lead; Extraction; Complications; Emergent surgical intervention ABBREVIATIONS ICD ¼ implantable cardioverter-defibrillator; SVC ¼ superior vena cava; TLE ¼ transvenous lead extraction (Heart Rhythm 2014;11:419–425) I 2014 Heart Rhythm Society. All rights reserved.
need to upgrade to new technology, increasing incidence of cardiovascular implantable electronic device infection,4,5 lead and device malfunction,6,7 and what has been termed a mismatch between patient and device longevity.8 It is estimated that the demand for transvenous lead extraction (TLE) has reached an annual extraction rate of 10,000– 15,000 leads worldwide.9 Chronically implanted transvenous pacemaker and implantable cardioverter-defibrillator (ICD) leads often develop fibrous tissue ingrowth at the interface between adjacent leads, venous endothelium, and endocardium and cannot be removed with simple traction alone. In many instances, extraction techniques using specialized tools are needed to free the leads. Because of the difficulty in extracting chronically implanted leads, major complications and even death may occur during TLE. In a comprehensive http://dx.doi.org/10.1016/j.hrthm.2013.12.004
420 review of prior TLE publications including 19,780 patients, the incidence of major complications was 1.8%, with 0.4% procedural mortality.10 Laceration of a major thoracic vessel and perforation of the myocardium are catastrophic complications that can occur during TLE. Either of these complications can result in rapid demise owing to exsanguination, hemothorax, and/ or cardiac tamponade. In some instances, these complications can be managed conservatively by administering intravenous fluids, vasopressors, and blood products or by performing pericardiocentesis or chest tube insertion. However, there is a subset of patients who will ultimately die of these injuries without emergent surgical or endovascular intervention. The outcomes of patients requiring emergent surgical or endovascular intervention for catastrophic complications during TLE have not been well characterized. The mortality associated with these events is substantial and unfortunately even immediate and aggressive attempts to rescue these patients may not succeed. Our objectives were to evaluate the incidence of these complications, to describe the injuries, and to review patient management and outcomes.
Methods We retrospectively analyzed consecutive patients who underwent TLE of pacemaker and ICD leads at the Cleveland Clinic between August 1996 and September 2012. Demographic and procedural data were obtained from the electronic medical records and prospectively collected institutional databases. The study was approved by the Institutional Review Board of the Cleveland Clinic. Lead extraction was defined, in accordance with the 2009 Heart Rhythm Society TLE consensus statement, as removal of a lead implanted for more than 1 year, or requiring specialized extraction tools, or from a route other than the implant vein.9 Patients who did not meet these criteria were excluded. Major complications were outcomes related to the procedure that were life-threatening, resulted in death, caused persistent or significant disability, or required significant surgical intervention to prevent any of the prior outcomes. Catastrophic complications requiring emergent surgical or endovascular intervention were outcomes in which emergent surgical or endovascular intervention was undertaken during or immediately after the TLE procedure for a complication arising from the procedure. Interventions for complications diagnosed after the TLE procedure were not included, as these form a heterogeneous group with variable time course and less urgent management. The TLE procedures were performed in the electrophysiology laboratories at the Cleveland Clinic with a cardiothoracic surgeon on site and immediately available. Patients were prepped and draped for emergency sternotomy or thoracotomy, and the rooms were equipped with an “open chest” cart including all necessary equipment for immediate surgical intervention. Imaging modalities including
Heart Rhythm, Vol 11, No 3, March 2014 transthoracic, transesophageal, and intracardiac echocardiography were available and frequently used. Cardiopulmonary bypass equipment and a perfusionist were available by request but were not present in the laboratory. Invasive hemodynamic monitoring and large-bore venous access were uniform, and a temporary pacemaker wire was placed as indicated. Cases were routinely performed under general anesthesia. Traction was applied to all leads at the beginning of each case. If traction alone did not result in successful lead extraction, a powered sheath was used. Powered sheaths included the initial generation, SLS II, and GlideLight Laser Sheath (Spectranetics, Colorado Springs, CO); the Evolution mechanical dilator sheath (Cook Medical, Bloomington, IN); and the Perfecta electrosurgical dissection sheath (Cook Medical, Bloomington, IN). The GlideLight Laser Sheath was available for use in cases after June 2012. In some instances, mechanical sheaths, snares, and/or a femoral approach was needed for successful extraction. The clinical and procedural characteristics of patients who had catastrophic complications requiring emergent surgical or endovascular intervention were compared with patients who did not. An analysis was also performed to determine whether there were clinical or procedural characteristics associated with death when catastrophic complications requiring emergent surgical or endovascular intervention occurred. Cases in which the death was thought to be related to a procedural complication requiring emergent surgical or endovascular intervention were included in the analysis. Continuous data are presented as median (25th, 75th percentile); categorical data are presented as number (percentage). Continuous variables were compared by using the Mann-Whitney U test, and categorical variables were compared by using the χ2 test or the Fisher exact test, as appropriate. Data were collected and analyzed by using JMP Pro 9.0.0 software (SAS Institute Inc, Cary, NC). A P value of o.05 was considered statistically significant.
Results Between August 1996 and September 2012, 3258 consecutive patients underwent TLE procedures at the Cleveland Clinic. The median patient age was 67.0 (55.0, 76.1) years, and 2256 (69.2%) were men. A total of 5973 leads were extracted, of which 4436 [74.3%] were pacemaker and 1537 (25.7%) were ICD leads. The median lead implant duration was 4.9 (2.4, 8.4) years, and 2.0 (1.0, 2.0) leads were extracted per procedure. Pacemaker leads had longer median implant duration than did ICD leads (5.2 [2.4, 9.1] vs 4.2 [2.4, 6.3] years, respectively; P o .001). Powered sheaths were used in 2369 (72.7%) of the procedures. Fifty-eight (1.8%) patients had a major complication, and 13 (0.4% incidence) procedure-related (6 procedural, 7 postprocedure) deaths occurred. Catastrophic complications requiring emergent surgical or endovascular intervention occurred in 25 (0.8%) patients.
Brunner et al Lead Extraction Complications Requiring Surgery
Factors associated with catastrophic complications requiring emergent surgical or endovascular intervention The clinical and procedural characteristics of the patients who had catastrophic complications requiring emergent surgical or endovascular intervention are compared with patients who did not in Tables 1 and 2. A laser sheath was used in most cases (23 of 25 [92.0%]) in which catastrophic complications requiring emergent surgical or endovascular intervention occurred, of which 16 (70.0%) were 16 F and 7 (30.0%) were 14 F. An operator experience, including the total number of prior lead extraction procedures and the total number of prior leads extracted, did not have a significant association with the likelihood of catastrophic complications requiring emergent surgical or endovascular intervention occurring. When catastrophic complications requiring emergent surgical or endovascular intervention occurred, the procedural access site was most commonly left pectoral (19 of 25 [76.0%]), followed by right pectoral (4 of 25 [16.0%]) and both right and left pectoral (2 of 25 [8.0%]). Data for the procedural access site were not available for all the patients who did not experience catastrophic complications requiring emergent surgical or endovascular intervention, and an analysis to determine the relative risk of extraction from either pectoral site could not be performed. In our patient Table 1
421 population, the majority of extractions are from a left pectoral access site. Right ventricular pacemaker leads (15 of 25 [60%]), right ventricular dual-coil (5 of 25 [20.0%]) and single-coil (2 of 25 [8.0%]) ICD leads, and right atrial pacemaker leads (2 of 25 [8.0%]) were extracted when the events occurred. In 1 (4.0%) instance, the type of lead associated with the injury was not determined because the symptoms were not evident until after complete extraction of more than 1 lead. ICD leads were extracted in 10 of 25 (40%) of the patients with catastrophic complications requiring emergent surgical or endovascular intervention and were not always the lead associated with injury. There was no association between the type of lead extracted (eg, right ventricular pacemaker, ICD (any), or dual-coil ICD) and the occurrence of catastrophic complications requiring emergent surgical or endovascular intervention.
Location of injury The locations of the injuries that occurred in cases requiring emergent surgical or endovascular intervention were identified through surgical exploration or angiography and are summarized in Table 3. The superior vena cava (SVC) was most commonly (16 of 25 [64.0%]) the location of injury.
Clinical Characteristics (n ¼ 3258 patients)
Characteristic Patient demographic characteristics/medical history Age (y) Male sex Body mass index (kg/m2) Diabetes End-stage renal disease Coronary artery disease Prior open chest surgery Left ventricular ejection fraction (%) NYHA functional class III or IV Indication for lead extraction* Infection Lead malfunction Upgrade of preexisting system Other Lead data No. of leads extracted per procedure ICD lead (any) extracted during procedure Dual-coil ICD lead extracted during procedure Median lead age (y) Age of the oldest lead extracted (y) Combined age of leads extracted (y)
Patients requiring emergent surgical or endovascular intervention (n ¼ 25)
Patients not requiring emergent surgical or endovascular intervention (n ¼ 3233)
70.6 (51.1, 75.7) 16 (64.0) 26.5 (22.3, 33.8) 6 (24.0) 0 (0.0) 14 (56.0) 5 (20.0) 40.5 (20.0, 55.0) 3 (12.0)
67.0 (55.0, 76.1) 2240 (69.3) 27.0 (24.0, 31.0) 819 (25.3) 117 (3.6) 1710 (52.9) 1146 (35.4) 35.0 (25.0, 55.0) 669 (20.7)
.97 .55 .55 .88 .33 .74 .11 .59 .28
1370 (42.4) 1145 (35.4) 514 (15.9) 213 (6.6)‡
.87 .95 .99 .85
11 (44.0) 9 (36.0) 4 (16.0) 1 (4.0)† 2.0 (1.0, 2.5) 10 (40.0) 7 (28.0) 9.2 (5.0, 12.3) 9.7 (5.6, 12.7) 12.7 (9.3, 29.7)
2.0 (1.0, 2.0) 1334 (41.3) 1012 (31.3) 4.8 (2.6, 7.8) 5.2 (2.8, 8.7) 7.3 (3.8, 14.6)
P
.14 .90 .73 o.001 .0001 o.001
Continuous variables are presented as median (25th, 75th percentile); categorical variables are presented as n (%). ICD ¼ implantable cardioverter-defibrillator; NYHA ¼ New York Heart Association. * More than 1 indication for extraction was present in some patients. † The other indication for extraction was need for radiotherapy. ‡ Other indications for extraction included retained lead fragments (80 [2.5%]), venous stasis/occlusion (38 [1.2%]), no further indication for device therapy (35 [1.1%]), pain (18 [0.6%]), unknown (10 [0.3%]), lead recall (9 [0.3%]), lead perforation (6 [0.2%]), need for radiotherapy (6 [0.2%]) or magnetic resonance imaging (6 [0.2%]), and severe tricuspid regurgitation (5 [0.2%]). Retained lead fragments were encountered after cardiovascular surgery or failed prior transvenous extraction attempts.
422 Table 2
Heart Rhythm, Vol 11, No 3, March 2014 Procedural characteristics (n ¼ 3258 patients)
Characteristic
Patients requiring emergent surgical or endovascular intervention (n ¼ 25)
Patients not requiring emergent surgical or endovascular intervention (n ¼ 3233)
P
Length of procedure (min) Fluoroscopy time (min) Powered sheath (any) Laser sheath Evolution mechanical dilator sheath Electrosurgical dissection sheath Mechanical sheath Femoral approach
155.0 (122.5, 276.5) 13.3 (9.5, 31.2) 24 (96.0) 23 (92.0) 3 (12.0) 1 (4.0) 4 (16.0) 3 (12.0)
135.0 (100.0, 184.2) 11.5 (5.9, 21.8) 2345 (72.5) 2201 (68.1) 63 (2.0) 154 (4.7) 437 (13.5) 129 (4.0)
.02 .18 o.01 .02 o.001 .86 .72 .04
Continuous variables are presented as median (25th, 75th percentile); categorical variables are presented as n (%).
Nine of 16 (56.0%) SVC injuries occurred at the right atrial junction. The right atrial or ventricular myocardium (5 of 25 [20%]) and brachiocephalic or subclavian vasculature (4 of 25 [16%]) were less commonly the locations of injuries. The injury in all but one instance was either vascular laceration or cardiac perforation. In 1 case, an arteriovenous fistula was created between the left brachiocephalic vein and the brachiocephalic artery.
Detection of catastrophic complications In 11 (44.0%) cases in which catastrophic complications requiring emergent surgical or endovascular intervention occurred, signs were noted immediately after the removal of a completely extracted lead. Signs were noted during attempted lead extraction or minutes after complete lead extraction in 8 (32.0%) and 6 (24.0%) cases, respectively. Signs presented within 15 minutes in the 6 cases where a complication was not immediately apparent after complete lead extraction. The initial finding in most patients was hypotension (22 of 25 [88.0%]) or profuse bleeding in the pocket (3 of 25 [12.0%]). High pressure and oxygenated blood return from the venous sheath (1 of 25 [4.0%]) was noted in addition to hypotension in the patient with an arteriovenous fistula. Pericardial effusion (17 of 25 [68.0%]) and hemothorax (4 of 25 [16.0%]) were often identified in association with hypotension.
Management of catastrophic complications Intravenous fluids, vasopressor agents, and blood products were commonly administered to patients requiring emergent Table 3 Location of injury in cases requiring emergent surgical or endovascular intervention (n ¼ 25) Superior vena cava-right atrial junction 9 (36.0) Superior vena cava 5 (20.0) Right ventricle 3 (12.0) Superior vena cava-brachiocephalic vein junction 2 (8.0) Right atrial appendage 2 (8.0) Left brachiocephalic vein and brachiocephalic artery 1 (4.0) Left subclavian artery 1 (4.0) Left subclavian vein 1 (4.0) Left axillary artery 1 (4.0) Variables are presented as n (%).
surgical or endovascular intervention after hypotension was noted. Cardiopulmonary resuscitation was performed as needed. Many of the patients requiring emergent surgical or endovascular intervention were initially managed with pericardiocentesis (14 of 25 [56.0%]) or chest tube insertion (2 of 25 [8.0%]). One of the patients who required pericardiocentesis also underwent a subxiphoid pericardial window before sternotomy. There were patients in our extraction experience who required emergent periocardiocentesis (6 of 3258 [0.2%]) or chest tube insertion (2 of 3258 [0.1%]) but did not require emergent surgical or endovascular intervention owing to clinical stabilization. A total of 24 emergent pericardiocentesis (n ¼ 20) or chest tube insertions (n ¼ 4) were performed, and 16 of the 24 (66.6%) patients ended up requiring emergent surgical or endovascular intervention. Surgery (n ¼ 22) or endovascular intervention (n ¼ 3) was performed in the cardiothoracic or vascular surgery operating room in 14 (56.0%) cases and the electrophysiology laboratory in 11 (44.0%) cases. All the TLE procedures were performed in our electrophysiology laboratories, and patient stability was a major determinant of the location in which the operations or endovascular interventions were performed. Twenty patients required either sternotomy (n ¼ 18) or thoracotomy (n ¼ 2) for SVC laceration (n ¼ 15) and right atrial (n ¼ 2) or ventricular (n ¼ 3) perforation. Two patients required vascular repair at the procedural access site for either subclavian vein or artery laceration. Three patients were managed with an endovascular approach for SVC laceration, left axillary artery laceration, and left brachiocephalic vein and brachiocephalic artery fistula. Three SVC lacerations were repaired with a pericardial patch; all other surgical repairs were primary. Cardiopulmonary bypass was used in 10 cases (median pump time 81.0 [45.0, 116.0] minutes), and all were for SVC laceration. Cases involving cardiopulmonary bypass were performed in the cardiothoracic surgery operating room under normothermic conditions. In almost all cases, any remaining leads or lead fragments were removed during the subsequent surgeries. In 1 instance, there were lead remnants retained in the subclavian vein that could not be removed. In another instance, a decision was made to abandon a lead that had not been extracted. Of the 16 patients with SVC injury, 6 were too unstable to be transferred from the electrophysiology laboratory to an
Brunner et al Lead Extraction Complications Requiring Surgery operating room. Three of these patients underwent successful repair off-pump in the electrophysiology laboratory. Two patients died during attempted repair, one of which had a history of cardiac surgery and only a lateral thoracotomy was performed. One patient was managed successfully with an endovascular stent. The cases managed with an endovascular approach included a fistula between the left brachiocephalic vein and the brachiocephalic artery treated by deploying a 10 mm 38 mm iCAST (Atrium Medical Corporation, Hudson, NH) balloon expandable covered stent in the artery, a left axillary artery laceration treated by deploying an 8 mm 25 mm GORE VIABAHN Endoprosthesis (W.L. Gore & Associates, Inc, Flagstaff, AZ), and a SVC laceration treated with a 23 mm 33 mm GORE EXCLUDER Aortic Extender Endoprosthesis (W.L. Gore & Associates, Inc). The first 2 procedures were performed in the vascular surgery operating room and the last in the electrophysiology laboratory. These cases occurred from 2006 through 2012.
Outcomes when catastrophic complications requiring emergent surgical or endovascular intervention occurred The outcomes of patients who experienced catastrophic complications requiring emergent surgical or endovascular intervention are summarized in Table 4. In-hospital mortality was 36.0% (6 deaths in the electrophysiology laboratory or operating room and 3 deaths during the subsequent hospitalization). Sixteen (64.0%) patients were discharged alive. Two of the postoperative deaths occurred the first day after the procedure and 1 on day 13. The 2 patients who died on the first day after their procedure had poor neurologic function, and care was withdrawn at the request of their family members. The patient that died on day 13 had sepsis and multiorgan failure. Each of the deaths was directly related to procedural complications. Median length of stay for patients requiring emergent surgical or endovascular intervention who survived to discharge was 13.5 (10.8, 14.0) days, with a range of 4 to 80 days. Data for 22 patients were available 12 months after the procedures, and 12 (54.5%) were alive. Permanent physical disability was difficult to determine because many of the patients were referred for complex extractions and then did not follow with us after discharge. One of the survivors had anoxic brain injury and had permanent physical disability. An additional patient required tracheostomy for prolonged weaning from the ventilator and died at a longterm acute care facility on postoperative day 94. Among the other surviving patients, there were not any permanent disabilities that were apparent or documented at discharge.
Predictors of mortality when catastrophic complications requiring emergent surgical or endovascular intervention occurred Older age (74.4 [63.6, 83.1] vs 63.0 [45.3, 72.5] years; P ¼ .02) was associated with mortality if catastrophic complications requiring emergent surgical or endovascular
423 Table 4 Outcomes of patients requiring emergent surgical or endovascular intervention (n ¼ 25) Survival to discharge 16 (64.0) Death during procedure or hospitalization 9 (36.0) Death in electrophysiology laboratory 5 (20.0) Death in operating room 1 (4.0) Death during hospitalization 3 (12.0) Discharge to home 11 (44.0) Discharge to skilled nursing facility 3 (12.0) Discharge to LTAC for prolonged ventilator wean 1 (4.0) Discharge LTAC with anoxic brain injury 1 (4.0) Length of stay if survived (d) 13.5 (10.8, 14.0) Continuous variables are presented as median (25th, 75th percentile); categorical variables are presented as n (%). LTAC ¼ long-term acute care facility.
intervention occurred. Sex, body mass index, diabetes, endstage renal disease, coronary artery disease, prior open chest surgery, left ventricular ejection fraction, and New York Heart Association functional class were not associated with mortality. Extraction for infection was associated with increased mortality if catastrophic complications requiring emergent surgical or endovascular intervention occurred (7 of 11 [77.8%] with infection vs 2 of 14 [14.3%] with other indications for extraction died; P ¼ .01). The type of lead extracted when injury occurred, specialized extraction tools used, size of laser sheath, location of injury, timing and character of the initial signs of injury, initial management strategy (eg, pericardiocentesis or chest tube placement), and surgical or endovascular approach were not significantly associated with increased mortality among patients with catastrophic complications requiring emergent surgical or endovascular intervention. Mortality was higher if emergent surgical or endovascular intervention was performed in the electrophysiology laboratory than in an operating room (7 of 11 [63.6%] deceased vs 2 of 14 [14.3%] deceased; P ¼ .01).
Mortality related to TLE in our total experience In our total experience including 3258 patients, there were 13 (0.4% incidence) procedure-related deaths and 9 are represented in this analysis. The other deaths were due to pulseless electrical activity during the procedure (n ¼ 2), postprocedural complications related to occlusion of an aortobifemoral graft that had been accessed for hemodynamic monitoring during a case, and avulsion of a tricuspid valve leaflet resulting in severe tricuspid regurgitation, postprocedure cardiogenic shock, and death during the placement of a Tandem Heart percutaneous left ventricular assist device several days later. These deaths were excluded because emergent surgical or endovascular intervention was either not performed or did not occur during or immediately after the TLE procedures. The pulseless electrical activity associated deaths occurred during a TLE attempt in 1 instance and minutes after a successful TLE in the other instance.
Discussion We have presented our data describing patients with catastrophic complications requiring emergent surgical or
424 endovascular intervention from our experience in the TLE of nearly 6000 chronic endovascular pacemaker and defibrillator leads. Our objectives were to evaluate the incidence of these complications, to describe the injuries, and to review patient management and outcomes. Our incidence of major complications and procedural mortality were in line with previously published single- and multicenter TLE experiences.10,11 Catastrophic complications requiring emergent surgical or endovascular intervention occurred in 25 (0.8%) patients, and the associated in-hospital mortality was 36.0% (9 deaths). Sixty-four percent of the patients were discharged from the hospital after the events. The outcomes of patients requiring emergent surgical or endovascular intervention for catastrophic complications occurring during TLE have not been well characterized. In a review of the US Food and Drug Administration’s Manufacturers and User Defined Experience database from 1995 to 2008, 62 patients requiring emergency surgical repair of venous lacerations and myocardial perforations during TLE were identified; the associated mortality was 44.0% (27 of 62 patients).12 Mortality was 50.0% (17 of 34 patients) if the injury occurred during laser extraction. The SVC was the site of injury in the majority of cases. In a single-center retrospective series of 112 consecutive laser-assisted lead extractions from 2002 to 2008, emergent surgical intervention was required for 4 (3.6%) patients (2 SVC lacerations, 1 subclavian vein laceration, and 1 right atrial perforation) and 3 of the patients died (75% mortality).13 In a retrospective analysis of 2201 patients undergoing TLE of 2274 ICD leads at 9 high-volume centers between 2000 and 2011, 4 cases (0.2% incidence) of pericardial tamponade requiring urgent/emergent thoracotomy/sternotomy for treatment were identified.14 In these patients, the injuries included 3 SVC lacerations and a right atrial perforation. One (25.0%) of the patients subsequently died after emergent sternotomy. In our experience, injury occurred most often at the junction of the SVC and the right atrium in cases requiring emergent surgical or endovascular intervention. In all cases, signs were noted during lead extraction or within 15 minutes later. Bleeding was more likely into the pericardial space than into the right pleural space, indicating that most injuries occurred below the pericardial reflection. If a patient required emergent pericardiocentesis or chest tube insertion, 66.6% (16 of 24) eventually required emergent surgical or endovascular intervention. Therefore, the cardiothoracic surgical team should be notified immediately upon recognition of a new pericardial or pleural effusion during TLE. Communication between the electrophysiologist and the surgeon is of utmost importance as the site of injury can be predicted on the basis of the location of the extraction tools when signs of injury are first noted. Longer lead implant duration, powered sheaths, and need for femoral approach were significantly associated with catastrophic complications requiring emergent surgical or endovascular intervention, reflecting the complexity of these
Heart Rhythm, Vol 11, No 3, March 2014 extractions. Procedural length was longer in cases requiring emergent surgical or endovascular intervention but was likely a reflection of the added time needed to diagnose and manage the complications. Variables associated with mortality when catastrophic complications requiring emergent surgical or endovascular intervention occurred included older patient age, lead extraction for infection, and surgery or endovascular intervention performed in the electrophysiology laboratory. The extractions were performed in our electrophysiology laboratories, and the latter association probably reflects the critical status of the patient, limiting the ability to be transported safely to the cardiothoracic or vascular surgery operating rooms. Conclusions regarding the ability to rescue a patient in one venue vs another cannot be drawn from these data, and we have not changed our practice on the basis of this analysis. As discussed above, our outcomes are in line with those reported at other centers, many of which perform their extractions in an operating room.10,11 In prior studies, body mass index o25 kg/m2,11 female sex,15–17 longer lead implant duration,17 elevated white blood cell count,18 dual-coil ICD leads,14 and laser lead extraction from both the right and the left extremity during the same procedure19 have been associated with major complications. Prior open chest surgery was protective for major complications in 1 study.18 In this analysis, body mass index, sex, and prior open chest surgery were not positively or negatively associated with catastrophic complications requiring emergent surgical or endovascular intervention. Right ventricular pacemaker leads were more often extracted than were single- or dual-coil ICD leads when catastrophic complications requiring emergent surgical or endovascular intervention occurred, which was likely related to longer implant duration in the pacemaker leads. To our knowledge, endovascular management of vascular injuries occurring during TLE has not been reported. There were no mortalities in these 3 patients, one of which included SVC laceration resulting in hemothorax. Further consideration of this management strategy is warranted, particularly for SVC injury above the pericardial reflection. It is plausible that a less invasive endovascular approach implemented rapidly could be associated with more favorable patient mortality and morbidity than does open chest surgery. As a result of our experience, we have created an “endovascular intervention” cart that is immediately available during our extraction cases.
Study limitations Our study has several limitations. The Cleveland Clinic is a high-volume tertiary referral center and outcomes may differ from centers with lower volume.9 It is possible that patients referred to our institution had indications for extraction, comorbidities, and perceived procedural complexity that are not typically encountered at less experienced centers. However, the baseline and lead characteristics of patients who did not experience catastrophic complications requiring emergent surgical or endovascular intervention were similar to
Brunner et al Lead Extraction Complications Requiring Surgery those represented in other contemporary lead extraction experiences. Our findings are also subject to the limitations inherent to retrospective studies, including the possibility of unknown confounders and bias in management strategy.
Conclusions Catastrophic complications requiring emergent surgical or endovascular intervention during TLE are uncommon but carry significant mortality. The SVC is the most frequent site for injuries. Despite high mortality, nearly two-thirds of these patients were rescued with immediate response and surgical or endovascular intervention.
References 1. Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 Guidelines for Device-Based Therapy of Cardiac Rhythm Abnormalities: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices) developed in collaboration with the American Association for Thoracic Surgery and Society of Thoracic Surgeons. J Am Coll Cardiol 2008;51:e1–e62. 2. Greenspon AJ, Patel JD, Lau E, et al. Trends in permanent pacemaker implantation in the United States 1993-2009: increasing complexity or patients and procedures. J Am Coll Cardiol 2012;60:1540–1545. 3. Camm J, Nisam S. Implantable cardioverter-defibrillator utilization. Europace 2011;13:448. 4. Voigt A, Shalaby A, Saba S. Continued rise in rates of cardiovascular implantable electronic device infections in the United States: temporal trends and causative insights. Pacing Clin Electrophysiol 2010;33:414–419. 5. Cabell CH, Heidenreich PA, Chu VH, et al. Increasing rates of cardiac device infections among Medicare beneficiaries: 1990-1999. Am Heart J 2004;147: 582–586. 6. Maisel WH. Pacemaker and ICD generator reliability: meta-analysis of device registries. JAMA 2006;295:1929–1934.
425 7. Liu J, Brumberg G, Rattan R, Jain S, Saba S. Class I recall of defibrillator leads: a comparison of the Sprint Fidelis and Riata families. Heart Rhythm 2012;9: 1251–1255. 8. Hauser RG. The growing mismatch between patient longevity and the service life of implantable cardioverter-defibrillators. J Am Coll Cardiol 2005;45: 2022–2025. 9. Wilkoff BL, Love CJ, Byrd CL, et al. Transvenous lead extraction: Heart Rhythm Society expert consensus on facilities, training, indications, and patient management: this document was endorsed by the American Heart Association (AHA). Heart Rhythm 2009;6:1085–1104. 10. Maytin M, Epstein LM, Henrikson CA. Lead extraction is preferred for lead revisions and system upgrades: when less is more. Circ Arrhythm Electrophysiol 2010;3:413–424; discussion 424. 11. Wazni O, Epstein LM, Carrillo RG, et al. Lead extraction in the contemporary setting: the LExICon study. An observational retrospective study of consecutive laser lead extractions. J Am Coll Cardiol 2010;55:579–586. 12. Hauser RG, Katsiyiannis WT, Gornick CC, Almquist AK, Kallinen LM. Deaths and cardiovascular injuries due to device-assisted implantable cardioverterdefibrillator and pacemaker lead extraction. Europace 2009;12:395–401. 13. Gaca JG, Lima B, Milano CA, et al. Laser-assisted extraction of pacemaker and defibrillator leads: the role of the cardiac surgeon. Ann Thorac Surg 2009;87: 1446–1450; discussion 1450–1441. 14. Epstein LM, Love CJ, Wilkoff BL, et al. Superior vena cava defibrillator coils make transvenous lead extraction more challenging and riskier. J Am Coll Cardiol 2013;61:987–989. 15. Byrd CL, Wilkoff BL, Love CJ, Sellers TD, Reiser C. Clinical study of the laser sheath for lead extraction: the total experience in the United States. Pacing Clin Electrophysiol 2002;25:804–808. 16. Byrd CL, Wilkoff BL, Love CJ, et al. Intravascular extraction of problematic or infected permanent pacemaker leads: 1994-1996. U.S. Extraction Database, MED Institute. Pacing Clin Electrophysiol 1999;22:1348–1357. 17. Kay GN, Brinker JA, Kawanishi DT, et al. Risks of spontaneous injury and extraction of an active fixation pacemaker lead: report of the Accufix Multicenter Clinical Study and Worldwide Registry. Circulation 1999;100:2344–2352. 18. Agarwal SK, Kamireddy S, Nemec J, Voigt A, Saba S. Predictors of complications of endovascular chronic lead extractions from pacemakers and defibrillators: a single-operator experience. J Cardiovasc Electrophysiol 2009;20:171–175. 19. Roux JF, Page P, Dubuc M, et al. Laser lead extraction: predictors of success and complications. Pacing Clin Electrophysiol 2007;30:214–220.
Introduction Cardiovascular implantable electronic device utilization has grown substantially in response to broadening evidencebased indications.1–3 In tandem, there has been a growing need for system revisions and lead extraction, driven by the Dr Saliba has received speaker’s honoraria from Boston Scientific, Medtronic, and St Jude Medical. Dr Lindsay is a member of the CardioInsight Technologies Physician Advisory Board. Dr Wilkoff has received speaker’s honoraria from Boston Scientific and St Jude Medical; he is also a member of the St Jude Medical Physician Advisory Board and Spectranetics Physician Advisory Board. Dr Tarakji has received speaker’s honoraria from HMP Communications. All authors with disclosures reported compensation less then $10,000 per year. Address reprint requests and correspondence: Dr Khaldoun G. Tarakji, Section of Cardiac Pacing and Electrophysiology, Heart and Vascular Institute, Cleveland Clinic, 9500 Euclid Avenue J2-2, Cleveland, OH 44195. E-mail address: [email protected].
1547-5271/$-see front matter B 2014 Heart Rhythm Society. All rights reserved.
(n ¼ 18) or thoracotomy (n ¼ 2) for superior vena cava laceration (n ¼ 15) and right atrial (n ¼ 2) or ventricular (n ¼ 3) perforation. Two patients required vascular repair at the procedural access site for either subclavian vein or artery laceration. Three patients were managed with an endovascular approach for superior vena cava laceration, left axillary artery laceration, and brachiocephalic vein and artery fistula. In-hospital mortality was 36.0% (6 procedural/operative deaths and 3 deaths during the same hospitalization). CONCLUSIONS Major vascular injury or cardiac perforation requiring emergent surgical or endovascular intervention during TLE is uncommon but carries significant in-hospital mortality. Despite high mortality, nearly two-thirds of these patients were rescued with immediate response and surgical or endovascular intervention. KEYWORDS Implantable cardioverter-defibrillator; Pacemaker; Lead; Extraction; Complications; Emergent surgical intervention ABBREVIATIONS ICD ¼ implantable cardioverter-defibrillator; SVC ¼ superior vena cava; TLE ¼ transvenous lead extraction (Heart Rhythm 2014;11:419–425) I 2014 Heart Rhythm Society. All rights reserved.
need to upgrade to new technology, increasing incidence of cardiovascular implantable electronic device infection,4,5 lead and device malfunction,6,7 and what has been termed a mismatch between patient and device longevity.8 It is estimated that the demand for transvenous lead extraction (TLE) has reached an annual extraction rate of 10,000– 15,000 leads worldwide.9 Chronically implanted transvenous pacemaker and implantable cardioverter-defibrillator (ICD) leads often develop fibrous tissue ingrowth at the interface between adjacent leads, venous endothelium, and endocardium and cannot be removed with simple traction alone. In many instances, extraction techniques using specialized tools are needed to free the leads. Because of the difficulty in extracting chronically implanted leads, major complications and even death may occur during TLE. In a comprehensive http://dx.doi.org/10.1016/j.hrthm.2013.12.004
420 review of prior TLE publications including 19,780 patients, the incidence of major complications was 1.8%, with 0.4% procedural mortality.10 Laceration of a major thoracic vessel and perforation of the myocardium are catastrophic complications that can occur during TLE. Either of these complications can result in rapid demise owing to exsanguination, hemothorax, and/ or cardiac tamponade. In some instances, these complications can be managed conservatively by administering intravenous fluids, vasopressors, and blood products or by performing pericardiocentesis or chest tube insertion. However, there is a subset of patients who will ultimately die of these injuries without emergent surgical or endovascular intervention. The outcomes of patients requiring emergent surgical or endovascular intervention for catastrophic complications during TLE have not been well characterized. The mortality associated with these events is substantial and unfortunately even immediate and aggressive attempts to rescue these patients may not succeed. Our objectives were to evaluate the incidence of these complications, to describe the injuries, and to review patient management and outcomes.
Methods We retrospectively analyzed consecutive patients who underwent TLE of pacemaker and ICD leads at the Cleveland Clinic between August 1996 and September 2012. Demographic and procedural data were obtained from the electronic medical records and prospectively collected institutional databases. The study was approved by the Institutional Review Board of the Cleveland Clinic. Lead extraction was defined, in accordance with the 2009 Heart Rhythm Society TLE consensus statement, as removal of a lead implanted for more than 1 year, or requiring specialized extraction tools, or from a route other than the implant vein.9 Patients who did not meet these criteria were excluded. Major complications were outcomes related to the procedure that were life-threatening, resulted in death, caused persistent or significant disability, or required significant surgical intervention to prevent any of the prior outcomes. Catastrophic complications requiring emergent surgical or endovascular intervention were outcomes in which emergent surgical or endovascular intervention was undertaken during or immediately after the TLE procedure for a complication arising from the procedure. Interventions for complications diagnosed after the TLE procedure were not included, as these form a heterogeneous group with variable time course and less urgent management. The TLE procedures were performed in the electrophysiology laboratories at the Cleveland Clinic with a cardiothoracic surgeon on site and immediately available. Patients were prepped and draped for emergency sternotomy or thoracotomy, and the rooms were equipped with an “open chest” cart including all necessary equipment for immediate surgical intervention. Imaging modalities including
Heart Rhythm, Vol 11, No 3, March 2014 transthoracic, transesophageal, and intracardiac echocardiography were available and frequently used. Cardiopulmonary bypass equipment and a perfusionist were available by request but were not present in the laboratory. Invasive hemodynamic monitoring and large-bore venous access were uniform, and a temporary pacemaker wire was placed as indicated. Cases were routinely performed under general anesthesia. Traction was applied to all leads at the beginning of each case. If traction alone did not result in successful lead extraction, a powered sheath was used. Powered sheaths included the initial generation, SLS II, and GlideLight Laser Sheath (Spectranetics, Colorado Springs, CO); the Evolution mechanical dilator sheath (Cook Medical, Bloomington, IN); and the Perfecta electrosurgical dissection sheath (Cook Medical, Bloomington, IN). The GlideLight Laser Sheath was available for use in cases after June 2012. In some instances, mechanical sheaths, snares, and/or a femoral approach was needed for successful extraction. The clinical and procedural characteristics of patients who had catastrophic complications requiring emergent surgical or endovascular intervention were compared with patients who did not. An analysis was also performed to determine whether there were clinical or procedural characteristics associated with death when catastrophic complications requiring emergent surgical or endovascular intervention occurred. Cases in which the death was thought to be related to a procedural complication requiring emergent surgical or endovascular intervention were included in the analysis. Continuous data are presented as median (25th, 75th percentile); categorical data are presented as number (percentage). Continuous variables were compared by using the Mann-Whitney U test, and categorical variables were compared by using the χ2 test or the Fisher exact test, as appropriate. Data were collected and analyzed by using JMP Pro 9.0.0 software (SAS Institute Inc, Cary, NC). A P value of o.05 was considered statistically significant.
Results Between August 1996 and September 2012, 3258 consecutive patients underwent TLE procedures at the Cleveland Clinic. The median patient age was 67.0 (55.0, 76.1) years, and 2256 (69.2%) were men. A total of 5973 leads were extracted, of which 4436 [74.3%] were pacemaker and 1537 (25.7%) were ICD leads. The median lead implant duration was 4.9 (2.4, 8.4) years, and 2.0 (1.0, 2.0) leads were extracted per procedure. Pacemaker leads had longer median implant duration than did ICD leads (5.2 [2.4, 9.1] vs 4.2 [2.4, 6.3] years, respectively; P o .001). Powered sheaths were used in 2369 (72.7%) of the procedures. Fifty-eight (1.8%) patients had a major complication, and 13 (0.4% incidence) procedure-related (6 procedural, 7 postprocedure) deaths occurred. Catastrophic complications requiring emergent surgical or endovascular intervention occurred in 25 (0.8%) patients.
Brunner et al Lead Extraction Complications Requiring Surgery
Factors associated with catastrophic complications requiring emergent surgical or endovascular intervention The clinical and procedural characteristics of the patients who had catastrophic complications requiring emergent surgical or endovascular intervention are compared with patients who did not in Tables 1 and 2. A laser sheath was used in most cases (23 of 25 [92.0%]) in which catastrophic complications requiring emergent surgical or endovascular intervention occurred, of which 16 (70.0%) were 16 F and 7 (30.0%) were 14 F. An operator experience, including the total number of prior lead extraction procedures and the total number of prior leads extracted, did not have a significant association with the likelihood of catastrophic complications requiring emergent surgical or endovascular intervention occurring. When catastrophic complications requiring emergent surgical or endovascular intervention occurred, the procedural access site was most commonly left pectoral (19 of 25 [76.0%]), followed by right pectoral (4 of 25 [16.0%]) and both right and left pectoral (2 of 25 [8.0%]). Data for the procedural access site were not available for all the patients who did not experience catastrophic complications requiring emergent surgical or endovascular intervention, and an analysis to determine the relative risk of extraction from either pectoral site could not be performed. In our patient Table 1
421 population, the majority of extractions are from a left pectoral access site. Right ventricular pacemaker leads (15 of 25 [60%]), right ventricular dual-coil (5 of 25 [20.0%]) and single-coil (2 of 25 [8.0%]) ICD leads, and right atrial pacemaker leads (2 of 25 [8.0%]) were extracted when the events occurred. In 1 (4.0%) instance, the type of lead associated with the injury was not determined because the symptoms were not evident until after complete extraction of more than 1 lead. ICD leads were extracted in 10 of 25 (40%) of the patients with catastrophic complications requiring emergent surgical or endovascular intervention and were not always the lead associated with injury. There was no association between the type of lead extracted (eg, right ventricular pacemaker, ICD (any), or dual-coil ICD) and the occurrence of catastrophic complications requiring emergent surgical or endovascular intervention.
Location of injury The locations of the injuries that occurred in cases requiring emergent surgical or endovascular intervention were identified through surgical exploration or angiography and are summarized in Table 3. The superior vena cava (SVC) was most commonly (16 of 25 [64.0%]) the location of injury.
Clinical Characteristics (n ¼ 3258 patients)
Characteristic Patient demographic characteristics/medical history Age (y) Male sex Body mass index (kg/m2) Diabetes End-stage renal disease Coronary artery disease Prior open chest surgery Left ventricular ejection fraction (%) NYHA functional class III or IV Indication for lead extraction* Infection Lead malfunction Upgrade of preexisting system Other Lead data No. of leads extracted per procedure ICD lead (any) extracted during procedure Dual-coil ICD lead extracted during procedure Median lead age (y) Age of the oldest lead extracted (y) Combined age of leads extracted (y)
Patients requiring emergent surgical or endovascular intervention (n ¼ 25)
Patients not requiring emergent surgical or endovascular intervention (n ¼ 3233)
70.6 (51.1, 75.7) 16 (64.0) 26.5 (22.3, 33.8) 6 (24.0) 0 (0.0) 14 (56.0) 5 (20.0) 40.5 (20.0, 55.0) 3 (12.0)
67.0 (55.0, 76.1) 2240 (69.3) 27.0 (24.0, 31.0) 819 (25.3) 117 (3.6) 1710 (52.9) 1146 (35.4) 35.0 (25.0, 55.0) 669 (20.7)
.97 .55 .55 .88 .33 .74 .11 .59 .28
1370 (42.4) 1145 (35.4) 514 (15.9) 213 (6.6)‡
.87 .95 .99 .85
11 (44.0) 9 (36.0) 4 (16.0) 1 (4.0)† 2.0 (1.0, 2.5) 10 (40.0) 7 (28.0) 9.2 (5.0, 12.3) 9.7 (5.6, 12.7) 12.7 (9.3, 29.7)
2.0 (1.0, 2.0) 1334 (41.3) 1012 (31.3) 4.8 (2.6, 7.8) 5.2 (2.8, 8.7) 7.3 (3.8, 14.6)
P
.14 .90 .73 o.001 .0001 o.001
Continuous variables are presented as median (25th, 75th percentile); categorical variables are presented as n (%). ICD ¼ implantable cardioverter-defibrillator; NYHA ¼ New York Heart Association. * More than 1 indication for extraction was present in some patients. † The other indication for extraction was need for radiotherapy. ‡ Other indications for extraction included retained lead fragments (80 [2.5%]), venous stasis/occlusion (38 [1.2%]), no further indication for device therapy (35 [1.1%]), pain (18 [0.6%]), unknown (10 [0.3%]), lead recall (9 [0.3%]), lead perforation (6 [0.2%]), need for radiotherapy (6 [0.2%]) or magnetic resonance imaging (6 [0.2%]), and severe tricuspid regurgitation (5 [0.2%]). Retained lead fragments were encountered after cardiovascular surgery or failed prior transvenous extraction attempts.
422 Table 2
Heart Rhythm, Vol 11, No 3, March 2014 Procedural characteristics (n ¼ 3258 patients)
Characteristic
Patients requiring emergent surgical or endovascular intervention (n ¼ 25)
Patients not requiring emergent surgical or endovascular intervention (n ¼ 3233)
P
Length of procedure (min) Fluoroscopy time (min) Powered sheath (any) Laser sheath Evolution mechanical dilator sheath Electrosurgical dissection sheath Mechanical sheath Femoral approach
155.0 (122.5, 276.5) 13.3 (9.5, 31.2) 24 (96.0) 23 (92.0) 3 (12.0) 1 (4.0) 4 (16.0) 3 (12.0)
135.0 (100.0, 184.2) 11.5 (5.9, 21.8) 2345 (72.5) 2201 (68.1) 63 (2.0) 154 (4.7) 437 (13.5) 129 (4.0)
.02 .18 o.01 .02 o.001 .86 .72 .04
Continuous variables are presented as median (25th, 75th percentile); categorical variables are presented as n (%).
Nine of 16 (56.0%) SVC injuries occurred at the right atrial junction. The right atrial or ventricular myocardium (5 of 25 [20%]) and brachiocephalic or subclavian vasculature (4 of 25 [16%]) were less commonly the locations of injuries. The injury in all but one instance was either vascular laceration or cardiac perforation. In 1 case, an arteriovenous fistula was created between the left brachiocephalic vein and the brachiocephalic artery.
Detection of catastrophic complications In 11 (44.0%) cases in which catastrophic complications requiring emergent surgical or endovascular intervention occurred, signs were noted immediately after the removal of a completely extracted lead. Signs were noted during attempted lead extraction or minutes after complete lead extraction in 8 (32.0%) and 6 (24.0%) cases, respectively. Signs presented within 15 minutes in the 6 cases where a complication was not immediately apparent after complete lead extraction. The initial finding in most patients was hypotension (22 of 25 [88.0%]) or profuse bleeding in the pocket (3 of 25 [12.0%]). High pressure and oxygenated blood return from the venous sheath (1 of 25 [4.0%]) was noted in addition to hypotension in the patient with an arteriovenous fistula. Pericardial effusion (17 of 25 [68.0%]) and hemothorax (4 of 25 [16.0%]) were often identified in association with hypotension.
Management of catastrophic complications Intravenous fluids, vasopressor agents, and blood products were commonly administered to patients requiring emergent Table 3 Location of injury in cases requiring emergent surgical or endovascular intervention (n ¼ 25) Superior vena cava-right atrial junction 9 (36.0) Superior vena cava 5 (20.0) Right ventricle 3 (12.0) Superior vena cava-brachiocephalic vein junction 2 (8.0) Right atrial appendage 2 (8.0) Left brachiocephalic vein and brachiocephalic artery 1 (4.0) Left subclavian artery 1 (4.0) Left subclavian vein 1 (4.0) Left axillary artery 1 (4.0) Variables are presented as n (%).
surgical or endovascular intervention after hypotension was noted. Cardiopulmonary resuscitation was performed as needed. Many of the patients requiring emergent surgical or endovascular intervention were initially managed with pericardiocentesis (14 of 25 [56.0%]) or chest tube insertion (2 of 25 [8.0%]). One of the patients who required pericardiocentesis also underwent a subxiphoid pericardial window before sternotomy. There were patients in our extraction experience who required emergent periocardiocentesis (6 of 3258 [0.2%]) or chest tube insertion (2 of 3258 [0.1%]) but did not require emergent surgical or endovascular intervention owing to clinical stabilization. A total of 24 emergent pericardiocentesis (n ¼ 20) or chest tube insertions (n ¼ 4) were performed, and 16 of the 24 (66.6%) patients ended up requiring emergent surgical or endovascular intervention. Surgery (n ¼ 22) or endovascular intervention (n ¼ 3) was performed in the cardiothoracic or vascular surgery operating room in 14 (56.0%) cases and the electrophysiology laboratory in 11 (44.0%) cases. All the TLE procedures were performed in our electrophysiology laboratories, and patient stability was a major determinant of the location in which the operations or endovascular interventions were performed. Twenty patients required either sternotomy (n ¼ 18) or thoracotomy (n ¼ 2) for SVC laceration (n ¼ 15) and right atrial (n ¼ 2) or ventricular (n ¼ 3) perforation. Two patients required vascular repair at the procedural access site for either subclavian vein or artery laceration. Three patients were managed with an endovascular approach for SVC laceration, left axillary artery laceration, and left brachiocephalic vein and brachiocephalic artery fistula. Three SVC lacerations were repaired with a pericardial patch; all other surgical repairs were primary. Cardiopulmonary bypass was used in 10 cases (median pump time 81.0 [45.0, 116.0] minutes), and all were for SVC laceration. Cases involving cardiopulmonary bypass were performed in the cardiothoracic surgery operating room under normothermic conditions. In almost all cases, any remaining leads or lead fragments were removed during the subsequent surgeries. In 1 instance, there were lead remnants retained in the subclavian vein that could not be removed. In another instance, a decision was made to abandon a lead that had not been extracted. Of the 16 patients with SVC injury, 6 were too unstable to be transferred from the electrophysiology laboratory to an
Brunner et al Lead Extraction Complications Requiring Surgery operating room. Three of these patients underwent successful repair off-pump in the electrophysiology laboratory. Two patients died during attempted repair, one of which had a history of cardiac surgery and only a lateral thoracotomy was performed. One patient was managed successfully with an endovascular stent. The cases managed with an endovascular approach included a fistula between the left brachiocephalic vein and the brachiocephalic artery treated by deploying a 10 mm 38 mm iCAST (Atrium Medical Corporation, Hudson, NH) balloon expandable covered stent in the artery, a left axillary artery laceration treated by deploying an 8 mm 25 mm GORE VIABAHN Endoprosthesis (W.L. Gore & Associates, Inc, Flagstaff, AZ), and a SVC laceration treated with a 23 mm 33 mm GORE EXCLUDER Aortic Extender Endoprosthesis (W.L. Gore & Associates, Inc). The first 2 procedures were performed in the vascular surgery operating room and the last in the electrophysiology laboratory. These cases occurred from 2006 through 2012.
Outcomes when catastrophic complications requiring emergent surgical or endovascular intervention occurred The outcomes of patients who experienced catastrophic complications requiring emergent surgical or endovascular intervention are summarized in Table 4. In-hospital mortality was 36.0% (6 deaths in the electrophysiology laboratory or operating room and 3 deaths during the subsequent hospitalization). Sixteen (64.0%) patients were discharged alive. Two of the postoperative deaths occurred the first day after the procedure and 1 on day 13. The 2 patients who died on the first day after their procedure had poor neurologic function, and care was withdrawn at the request of their family members. The patient that died on day 13 had sepsis and multiorgan failure. Each of the deaths was directly related to procedural complications. Median length of stay for patients requiring emergent surgical or endovascular intervention who survived to discharge was 13.5 (10.8, 14.0) days, with a range of 4 to 80 days. Data for 22 patients were available 12 months after the procedures, and 12 (54.5%) were alive. Permanent physical disability was difficult to determine because many of the patients were referred for complex extractions and then did not follow with us after discharge. One of the survivors had anoxic brain injury and had permanent physical disability. An additional patient required tracheostomy for prolonged weaning from the ventilator and died at a longterm acute care facility on postoperative day 94. Among the other surviving patients, there were not any permanent disabilities that were apparent or documented at discharge.
Predictors of mortality when catastrophic complications requiring emergent surgical or endovascular intervention occurred Older age (74.4 [63.6, 83.1] vs 63.0 [45.3, 72.5] years; P ¼ .02) was associated with mortality if catastrophic complications requiring emergent surgical or endovascular
423 Table 4 Outcomes of patients requiring emergent surgical or endovascular intervention (n ¼ 25) Survival to discharge 16 (64.0) Death during procedure or hospitalization 9 (36.0) Death in electrophysiology laboratory 5 (20.0) Death in operating room 1 (4.0) Death during hospitalization 3 (12.0) Discharge to home 11 (44.0) Discharge to skilled nursing facility 3 (12.0) Discharge to LTAC for prolonged ventilator wean 1 (4.0) Discharge LTAC with anoxic brain injury 1 (4.0) Length of stay if survived (d) 13.5 (10.8, 14.0) Continuous variables are presented as median (25th, 75th percentile); categorical variables are presented as n (%). LTAC ¼ long-term acute care facility.
intervention occurred. Sex, body mass index, diabetes, endstage renal disease, coronary artery disease, prior open chest surgery, left ventricular ejection fraction, and New York Heart Association functional class were not associated with mortality. Extraction for infection was associated with increased mortality if catastrophic complications requiring emergent surgical or endovascular intervention occurred (7 of 11 [77.8%] with infection vs 2 of 14 [14.3%] with other indications for extraction died; P ¼ .01). The type of lead extracted when injury occurred, specialized extraction tools used, size of laser sheath, location of injury, timing and character of the initial signs of injury, initial management strategy (eg, pericardiocentesis or chest tube placement), and surgical or endovascular approach were not significantly associated with increased mortality among patients with catastrophic complications requiring emergent surgical or endovascular intervention. Mortality was higher if emergent surgical or endovascular intervention was performed in the electrophysiology laboratory than in an operating room (7 of 11 [63.6%] deceased vs 2 of 14 [14.3%] deceased; P ¼ .01).
Mortality related to TLE in our total experience In our total experience including 3258 patients, there were 13 (0.4% incidence) procedure-related deaths and 9 are represented in this analysis. The other deaths were due to pulseless electrical activity during the procedure (n ¼ 2), postprocedural complications related to occlusion of an aortobifemoral graft that had been accessed for hemodynamic monitoring during a case, and avulsion of a tricuspid valve leaflet resulting in severe tricuspid regurgitation, postprocedure cardiogenic shock, and death during the placement of a Tandem Heart percutaneous left ventricular assist device several days later. These deaths were excluded because emergent surgical or endovascular intervention was either not performed or did not occur during or immediately after the TLE procedures. The pulseless electrical activity associated deaths occurred during a TLE attempt in 1 instance and minutes after a successful TLE in the other instance.
Discussion We have presented our data describing patients with catastrophic complications requiring emergent surgical or
424 endovascular intervention from our experience in the TLE of nearly 6000 chronic endovascular pacemaker and defibrillator leads. Our objectives were to evaluate the incidence of these complications, to describe the injuries, and to review patient management and outcomes. Our incidence of major complications and procedural mortality were in line with previously published single- and multicenter TLE experiences.10,11 Catastrophic complications requiring emergent surgical or endovascular intervention occurred in 25 (0.8%) patients, and the associated in-hospital mortality was 36.0% (9 deaths). Sixty-four percent of the patients were discharged from the hospital after the events. The outcomes of patients requiring emergent surgical or endovascular intervention for catastrophic complications occurring during TLE have not been well characterized. In a review of the US Food and Drug Administration’s Manufacturers and User Defined Experience database from 1995 to 2008, 62 patients requiring emergency surgical repair of venous lacerations and myocardial perforations during TLE were identified; the associated mortality was 44.0% (27 of 62 patients).12 Mortality was 50.0% (17 of 34 patients) if the injury occurred during laser extraction. The SVC was the site of injury in the majority of cases. In a single-center retrospective series of 112 consecutive laser-assisted lead extractions from 2002 to 2008, emergent surgical intervention was required for 4 (3.6%) patients (2 SVC lacerations, 1 subclavian vein laceration, and 1 right atrial perforation) and 3 of the patients died (75% mortality).13 In a retrospective analysis of 2201 patients undergoing TLE of 2274 ICD leads at 9 high-volume centers between 2000 and 2011, 4 cases (0.2% incidence) of pericardial tamponade requiring urgent/emergent thoracotomy/sternotomy for treatment were identified.14 In these patients, the injuries included 3 SVC lacerations and a right atrial perforation. One (25.0%) of the patients subsequently died after emergent sternotomy. In our experience, injury occurred most often at the junction of the SVC and the right atrium in cases requiring emergent surgical or endovascular intervention. In all cases, signs were noted during lead extraction or within 15 minutes later. Bleeding was more likely into the pericardial space than into the right pleural space, indicating that most injuries occurred below the pericardial reflection. If a patient required emergent pericardiocentesis or chest tube insertion, 66.6% (16 of 24) eventually required emergent surgical or endovascular intervention. Therefore, the cardiothoracic surgical team should be notified immediately upon recognition of a new pericardial or pleural effusion during TLE. Communication between the electrophysiologist and the surgeon is of utmost importance as the site of injury can be predicted on the basis of the location of the extraction tools when signs of injury are first noted. Longer lead implant duration, powered sheaths, and need for femoral approach were significantly associated with catastrophic complications requiring emergent surgical or endovascular intervention, reflecting the complexity of these
Heart Rhythm, Vol 11, No 3, March 2014 extractions. Procedural length was longer in cases requiring emergent surgical or endovascular intervention but was likely a reflection of the added time needed to diagnose and manage the complications. Variables associated with mortality when catastrophic complications requiring emergent surgical or endovascular intervention occurred included older patient age, lead extraction for infection, and surgery or endovascular intervention performed in the electrophysiology laboratory. The extractions were performed in our electrophysiology laboratories, and the latter association probably reflects the critical status of the patient, limiting the ability to be transported safely to the cardiothoracic or vascular surgery operating rooms. Conclusions regarding the ability to rescue a patient in one venue vs another cannot be drawn from these data, and we have not changed our practice on the basis of this analysis. As discussed above, our outcomes are in line with those reported at other centers, many of which perform their extractions in an operating room.10,11 In prior studies, body mass index o25 kg/m2,11 female sex,15–17 longer lead implant duration,17 elevated white blood cell count,18 dual-coil ICD leads,14 and laser lead extraction from both the right and the left extremity during the same procedure19 have been associated with major complications. Prior open chest surgery was protective for major complications in 1 study.18 In this analysis, body mass index, sex, and prior open chest surgery were not positively or negatively associated with catastrophic complications requiring emergent surgical or endovascular intervention. Right ventricular pacemaker leads were more often extracted than were single- or dual-coil ICD leads when catastrophic complications requiring emergent surgical or endovascular intervention occurred, which was likely related to longer implant duration in the pacemaker leads. To our knowledge, endovascular management of vascular injuries occurring during TLE has not been reported. There were no mortalities in these 3 patients, one of which included SVC laceration resulting in hemothorax. Further consideration of this management strategy is warranted, particularly for SVC injury above the pericardial reflection. It is plausible that a less invasive endovascular approach implemented rapidly could be associated with more favorable patient mortality and morbidity than does open chest surgery. As a result of our experience, we have created an “endovascular intervention” cart that is immediately available during our extraction cases.
Study limitations Our study has several limitations. The Cleveland Clinic is a high-volume tertiary referral center and outcomes may differ from centers with lower volume.9 It is possible that patients referred to our institution had indications for extraction, comorbidities, and perceived procedural complexity that are not typically encountered at less experienced centers. However, the baseline and lead characteristics of patients who did not experience catastrophic complications requiring emergent surgical or endovascular intervention were similar to
Brunner et al Lead Extraction Complications Requiring Surgery those represented in other contemporary lead extraction experiences. Our findings are also subject to the limitations inherent to retrospective studies, including the possibility of unknown confounders and bias in management strategy.
Conclusions Catastrophic complications requiring emergent surgical or endovascular intervention during TLE are uncommon but carry significant mortality. The SVC is the most frequent site for injuries. Despite high mortality, nearly two-thirds of these patients were rescued with immediate response and surgical or endovascular intervention.
References 1. Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 Guidelines for Device-Based Therapy of Cardiac Rhythm Abnormalities: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implantation of Cardiac Pacemakers and Antiarrhythmia Devices) developed in collaboration with the American Association for Thoracic Surgery and Society of Thoracic Surgeons. J Am Coll Cardiol 2008;51:e1–e62. 2. Greenspon AJ, Patel JD, Lau E, et al. Trends in permanent pacemaker implantation in the United States 1993-2009: increasing complexity or patients and procedures. J Am Coll Cardiol 2012;60:1540–1545. 3. Camm J, Nisam S. Implantable cardioverter-defibrillator utilization. Europace 2011;13:448. 4. Voigt A, Shalaby A, Saba S. Continued rise in rates of cardiovascular implantable electronic device infections in the United States: temporal trends and causative insights. Pacing Clin Electrophysiol 2010;33:414–419. 5. Cabell CH, Heidenreich PA, Chu VH, et al. Increasing rates of cardiac device infections among Medicare beneficiaries: 1990-1999. Am Heart J 2004;147: 582–586. 6. Maisel WH. Pacemaker and ICD generator reliability: meta-analysis of device registries. JAMA 2006;295:1929–1934.
425 7. Liu J, Brumberg G, Rattan R, Jain S, Saba S. Class I recall of defibrillator leads: a comparison of the Sprint Fidelis and Riata families. Heart Rhythm 2012;9: 1251–1255. 8. Hauser RG. The growing mismatch between patient longevity and the service life of implantable cardioverter-defibrillators. J Am Coll Cardiol 2005;45: 2022–2025. 9. Wilkoff BL, Love CJ, Byrd CL, et al. Transvenous lead extraction: Heart Rhythm Society expert consensus on facilities, training, indications, and patient management: this document was endorsed by the American Heart Association (AHA). Heart Rhythm 2009;6:1085–1104. 10. Maytin M, Epstein LM, Henrikson CA. Lead extraction is preferred for lead revisions and system upgrades: when less is more. Circ Arrhythm Electrophysiol 2010;3:413–424; discussion 424. 11. Wazni O, Epstein LM, Carrillo RG, et al. Lead extraction in the contemporary setting: the LExICon study. An observational retrospective study of consecutive laser lead extractions. J Am Coll Cardiol 2010;55:579–586. 12. Hauser RG, Katsiyiannis WT, Gornick CC, Almquist AK, Kallinen LM. Deaths and cardiovascular injuries due to device-assisted implantable cardioverterdefibrillator and pacemaker lead extraction. Europace 2009;12:395–401. 13. Gaca JG, Lima B, Milano CA, et al. Laser-assisted extraction of pacemaker and defibrillator leads: the role of the cardiac surgeon. Ann Thorac Surg 2009;87: 1446–1450; discussion 1450–1441. 14. Epstein LM, Love CJ, Wilkoff BL, et al. Superior vena cava defibrillator coils make transvenous lead extraction more challenging and riskier. J Am Coll Cardiol 2013;61:987–989. 15. Byrd CL, Wilkoff BL, Love CJ, Sellers TD, Reiser C. Clinical study of the laser sheath for lead extraction: the total experience in the United States. Pacing Clin Electrophysiol 2002;25:804–808. 16. Byrd CL, Wilkoff BL, Love CJ, et al. Intravascular extraction of problematic or infected permanent pacemaker leads: 1994-1996. U.S. Extraction Database, MED Institute. Pacing Clin Electrophysiol 1999;22:1348–1357. 17. Kay GN, Brinker JA, Kawanishi DT, et al. Risks of spontaneous injury and extraction of an active fixation pacemaker lead: report of the Accufix Multicenter Clinical Study and Worldwide Registry. Circulation 1999;100:2344–2352. 18. Agarwal SK, Kamireddy S, Nemec J, Voigt A, Saba S. Predictors of complications of endovascular chronic lead extractions from pacemakers and defibrillators: a single-operator experience. J Cardiovasc Electrophysiol 2009;20:171–175. 19. Roux JF, Page P, Dubuc M, et al. Laser lead extraction: predictors of success and complications. Pacing Clin Electrophysiol 2007;30:214–220.