Author + information
- Received May 22, 2018
- Revision received July 16, 2018
- Accepted July 17, 2018
- Published online November 19, 2018.
- Seung-Jung Park, MD, PhDa,b,
- James L. Gentry III, MDa,
- Niraj Varma, MD, PhDa,
- Oussama Wazni, MDa,
- Khaldoun G. Tarakji, MD, MPHa,
- Anand Mehta, MDc,
- Stephanie Mick, MDd,
- Richard Grimm, DOa and
- Bruce L. Wilkoff, MDa,∗ ()
- aDepartment of Cardiovascular Medicine, Cleveland Clinic, Cleveland, Ohio
- bDepartment of Internal Medicine, Heart Vascular Stroke Institute, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
- cDepartment of Cardiothoracic Anesthesiology, Cleveland Clinic, Cleveland, Ohio
- dDepartment of and Cardiothoracic Surgery, Cleveland Clinic, Cleveland, Ohio
- ↵∗Address for correspondence:
Dr. Bruce L. Wilkoff, Department of Cardiovascular Medicine, Desk J2-2, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, Ohio 44195.
Objectives The aims of this study were to detect and quantify acute increases in tricuspid regurgitation (TR) severity following transvenous lead extraction (TLE) and to evaluate the associated risk factors.
Background Although established as a safe and effective method for lead removal, TLE is sometimes complicated by TR.
Methods In 208 consecutive patients undergoing TLE, acute changes in TR severity were assessed by transesophageal echocardiography. A significant acute TR increase (TRI) was defined as a ≥1 grade increase in TR severity and post-extraction TR severity that was moderate or greater.
Results Overall, 266 ventricular leads (mean lead age, 11.8 ± 7.3 years) were extracted from the 208 patients. A significant acute TRI was observed in 24 (11.5%) of these patients. Acute TRI was associated with longer lead implant duration, extraction of pacemaker rather than defibrillator leads, anatomic injury to the tricuspid valve (TV), and longer post-extraction hospital stays. Multivariate analysis yielded only lead implant duration as an independent predictor of TLE-related acute TRI (odds ratio: 1.05; 95% confidence interval: 1.01 to 1.11; p = 0.046). When the patients were divided into 4 subgroups according to quartiles of lead age, there was a graded elevation in the rates of acute TRI (p trend = 0.048) and TV injury (p trend = 0.009) with lead implant duration.
Conclusions Following TLE, TV damage and acute TRI were commonly detected by transesophageal echocardiography, particularly in patients with advanced lead age. Lead abandonment strategies, which prolong implantation duration of future leads requiring extraction, should consider the potential long-term deleterious impact on TV function.
Lead management and extraction are increasingly essential components of the comprehensive care of patients with cardiovascular implantable electronic devices (CIEDs). Factors influencing the increased frequency include a recent rapid rise in de novo CIED implantations and frequent system revisions or upgrades and lead-related problems such as CIED infections, lead failures, and lead safety alerts (1). Transvenous lead extraction (TLE) is established as a safe and effective method of lead removal with a high rate of procedural success and a low level of major complications (2–4). However, TLE procedures are potentially complicated by significant tricuspid valve (TV) injury with a subsequent increase in tricuspid regurgitation (TR). This occurs when the extensive fibrotic adhesions between leads implanted on a long basis and between the lead and the TV apparatus are disrupted (5–10).
There are limited data on the effect of TLE on TV function. However, because TLE procedures are frequently monitored by transesophageal echocardiography (TEE), a valuable opportunity exists to monitor TV function before and after the extraction procedure. The objectives of the present study were to investigate the rate of significant changes in TR severity detected by TEE following TLE procedures and to evaluate the relevant risk factors.
Demographic, clinical, device-related, and procedure-related data from consecutive patients who underwent TLE of pacing or defibrillator leads with TEE monitoring at the Cleveland Clinic in Cleveland, Ohio between July 2014 and December 2016 were retrieved from the prospectively collected procedural database and electronic medical record. TLE cases, as defined by the 2017 Heart Rhythm Society expert consensus statement on CIED lead management and extraction, were eligible for inclusion: either removal of lead(s) implanted for >1 year or lead removal requiring specialized extraction equipment not used during implantation (11). We included only patients who underwent extraction of at least 1 right ventricular (RV) pacing or defibrillator lead. Patients with: 1) no adequate pre-TLE or post-TLE echocardiographic data; 2) extraction of atrial leads alone; 3) extraction of proximal remnant of RV lead dissected above the TV; 4) aborted TLE procedures caused by other complications; or 5) combined TV repair operations were excluded. The study protocol was approved by the Institutional Review Board of the Cleveland Clinic, and the requirement for written informed consent was waived.
All patients underwent lead extraction while they were under general anesthesia in the hybrid cardiothoracic operating room equipped with cardiac catheterization laboratory cine/fluoroscopic equipment (Siemens Artis Q.Zen, Erlangen, Germany) at the Cleveland Clinic with cardiothoracic surgery immediately available. Patients were prepared for the procedure according to the published best practice protocol (12). This included TEE and invasive arterial pressure monitoring, large-bore femoral central venous access, and temporary ventricular pacing when indicated, as well as a stiff guidewire and 12-F femoral venous sheath for potential insertion of a compliant occlusive balloon in the superior vena cava since approval by the Food and Drug Administration. Leads were prepared to provide improved tensile properties, including locking stylet, suture, 1-tie, and bulldog, as previously described as needed for particular leads (13). Unless the leads were removed with trivial traction, 1 of several types of powered sheaths was used to complete the extraction: SLS II or GlideLight Excimer Laser Sheath (Spectranetics/Philips, Colorado Springs, Colorado), Perfecta Electrosurgical Dissection Sheath (Cook Medical, Bloomington, Indiana), Evolution RL Rotating Dilator Sheath (Cook Medical), or TightRail Rotating Dilator Sheath (Spectranetics/Philips). In some cases mechanical sheaths, snares, or a femoral workstation were required to complete the lead removal.
After induction of general anesthesia and intubation, a TEE probe was placed to evaluate the pre-operative status of pericardial effusion, assess the structural integrity and function of the cardiac valves including the TV, and document vegetations on the valves or leads when endocarditis was suspected. TEE was then used to monitor for procedure-related complications during and immediately after the extraction. A single certified cardiologist/echocardiographer (J.G.), blinded to the clinical procedural results, reviewed and graded the TEE data, including color Doppler images and final reports. Pre-TLE and post-TLE TR severity was graded as trivial, mild, moderate, or severe according to international guidelines (14). A significant acute tricuspid regurgitation increase (TRI) was defined as an increase of at least 1 grade with a post-extraction severity of moderate or more.
Continuous variables are presented as median (25th, 75th percentile), and categorical variables are given as numbers (percentages). Comparisons among groups were performed using Mann-Whitney U tests, chi-square tests, or the Fisher exact test, as appropriate. The study group was divided into 4 groups on the basis of quartiles of the lead implantation age of oldest RV lead: first quartile (n = 53), <7 years; second quartile (n = 51), 7 to 10 years; third quartile (n = 53), 10 to 15 years; fourth quartile (n = 51), ≥15 years. The incidences of TV injury and acute TRI in each quartile were calculated and tested for trend across the quartiles by using the linear-by-linear association test. Univariate and multivariate analyses were performed using logistic regression with the clinical and device- or procedure-related variables to assess the risk factors for the acute TRI and to calculate adjusted odds ratio (ORs) and their 95% confidence intervals (CIs). Variables with p < 0.05 in the univariate analysis were included in the multivariate analysis. Statistical analyses were performed using PASW Statistics version 23 software for Microsoft (SPSS Inc., Chicago, Illinois). All tests were 2-tailed, and p < 0.05 was considered statistically significant.
From July 2014 to December 2016, a total of 337 consecutive patients underwent TLE under the guidance of TEE in the hybrid cardiothoracic operating room. Among the 337 patients, we excluded 129 patients for the following reasons: incomplete TEE data (n = 118); extraction of atrial lead alone (n = 4) or proximal remnant of RV lead alone (n = 3); aborted procedures as a result of complications; superior vena cava tearing (n = 1) and arteriovenous fistula in the subclavian vein (n = 1); removal of lead by simple traction within 1 year after implantation (n = 1); or combined TV replacement operation (n = 1). Finally, a total of 208 patients were included in the final analysis (Figure 1). From these patients, 456 endocardial leads (2.2 ± 1.0 per patient) were extracted; 266 RV, 147 right atrial, and 43 cardiac venous leads. Mean and median implantation ages of the oldest RV lead were 11.8 ± 7.3 years and 10.0 years (7.0 to 15.1 years), respectively. The most common 2 indications for extraction were infection (n = 106, 51%) and lead malfunction (n = 86, 41%). The other indications for the TLE procedure included upgrade of a pre-existing device system (n = 4), need for magnetic resonance imaging (n = 3), recalled leads without overt malfunction at the time of generator replacement (n = 3), patient’s request for relief of anxiety and pain (n = 2), lead dislodgment (n = 2), the absence of devices that can accommodate leads with outdated connectors (n = 1), and need for radiation therapy (n = 1).
Incidence of acute TRI and comparison of patient groups
Of the 208 patients, significantly increased TR was observed in 24 (11.5%) patients after extraction (Figure 1). An emergency TV replacement operation was performed for 1 patient with new onset severe TR secondary to avulsion of a TV septal leaflet. The remaining 23 patients were medically managed with diuretic initiation or dose adjustments. The details of change in TR severity observed immediately after the procedure are depicted in Figures 2A and 2B. Table 1 shows that there were no significant differences in baseline clinical characteristics and the indications for the TLE between patients with and without acute TRI. However, patients with acute TRI were more likely to have had pacemakers instead of defibrillator leads removed than were patients without acute TRI (Table 2). In addition, both the implantation duration of the oldest ventricular lead and the combined duration of all RV leads were longer in the group with acute TRI than in the group without acute TRI. There was a trend toward a greater number of extracted ventricular leads and extraction tools used per patient in the acute TRI group. Regarding the involvement rates of vegetation on the TV and/or RV leads detected by TEE, both groups showed no significant differences. However, evidence of TV damage, such as TV leaflet avulsion, newly detected TV prolapse, chordal rupture, and flail chordae, was more frequently observed in the acute TRI group (p < 0.001). Patients with acute TRI had longer hospital stays after extraction compared with patients without acute TRI even when the patient requiring a TV operation was excluded from the analysis. When patients were divided into 2 subgroups depending on the presence or absence of device system infection as a TLE indication, patients with acute TRI tended to have a longer post-TLE hospital stay than did patients without acute TRI (Online Figure 1).
Risk factors for acute TRI
Univariate analysis demonstrated that extraction of a pacing lead instead of a defibrillator lead, number of extracted RV leads ≥3, use of ≥2 extraction tools, lead implantation age of the oldest RV lead, and combined age of extracted RV leads were significantly associated with acute TRI (Table 3). By multivariable analysis, however, implantation duration of the oldest RV lead was identified as an independent predictor for the development of acute TRI after the TLE procedure (OR: 1.05; 95% CI: 1.01 to 1.11; p = 0.046). When the patients were divided into 4 subgroups according to quartiles of lead implantation age (oldest RV lead), there was a graded elevation in the rates of both acute TRI and TV injury with lead age quartiles (Figure 3). However, when examining only the second and third quartiles of lead age, to minimize the confounding effect of lead age versus lead type on the rate of acute TRI, there was no significant difference in the lead age between pacemaker and implantable cardioverter-defibrillator leads (10.5 ± 2.4 years vs. 9.6 ± 1.7 years; p = 0.091) or in the rate of acute TRI (11.1% vs. 8.8%; p = 0.710).
Merits of the present study
This is a large and high-risk dataset evaluated for true TLE-related acute TR. Previously reported TLE data usually used definitions of TLE from patients with leads <1 year old and without the use of extraction tools. Therefore, many patients, from 14% to 52%, undergoing manual traction alone were included in the previous studies (5–9). The recently updated (2017) Heart Rhythm Society lead extraction consensus statement clarified and emphasized the definition of TLE (11). All 208 patients in our study group met this precise definition. Additionally, this cohort had the most complex patient-related characteristics, with longer mean lead age (11.8 ± 7.3 years) and greater number of extracted RV leads (1.3 ± 0.6 per patient) compared with previous studies. Therefore, these data likely better reflect a real-world assessment of acute TRI during a true TLE procedure in patients with longer dwell time and multiple CIED leads. Finally, the use of TEE allowed us to detect vegetations on the TV or CIED leads and evidence of TV injury with greater sensitivity.
Risk factors for TLE-related TR aggravation
In previous studies, the incidence of acute TRI following TLE was reported over a relatively wide range, from 3.5% to 15% (5–10). The wide disparity may be a result of variability in the definition of acute TRI in each of those studies. However, the rate of acute TRI could also be affected by differences in dwell time of the extracted leads. Endocardial ventricular leads implanted on a long-term basis progressively develop fibrotic attachments to TV apparatus, as documented by autopsy studies (15), open heart surgical data (16), and several case reports of TLE (17–20). Therefore, longer indwelling time could lead to greater degree of fibrotic adhesion and an increased risk of TV injury during the extraction procedure. All the previous studies with a mean lead age <5 years reported rates of acute TRI <10% (6–8,10). In the other 2 studies with a mean lead age >5 years, the rate rose to more than 10% (5,9). Our results are in line with this trend. The overall rate of acute TRI was 11.5% in our patients with a mean dwell time of 11.8 years. Moreover, there was a graded increase in the acute TRI rate according to the quartile of lead age (Figure 3). Interestingly, a subgroup of patients in the lowest quartile of lead age showed a mean dwell time of 4.4 ± 2.0 years and a 5.6% acute TRI rate. Actually, a longer implantation time was closely associated with acute TRI in univariate analyses of 1 prospective and 1 retrospective study (6,8). Our study showed that lead age was identified as an independent predictor in multivariate analysis as well.
Compared with pacing leads, defibrillator leads have a more complex design and a larger surface area, with a distal coil frequently placed close to TV. Therefore, fibrotic tissue growing into the grooves of defibrillator coil could lead to a greater TV apparatus damage during release of the fibrotic attachments and could increase post-procedural TR severity. However, most previous evaluations showed no significant differences between extraction of defibrillator leads and extraction of pacing leads in terms of the risk of TR aggravation (5,6,9). Surprisingly, extraction of pacing leads, instead of defibrillator leads, was significantly associated with post-TLE acute TRI in our univariate analysis. Coffey et al. (8) also showed, in their retrospective study incorporating 124 patients, a similar result with extraction of pacing leads instead of defibrillator leads predicting worsening of TR. Lead implantation age was not considered in comparisons among lead types in most previous studies (6–10), and these unexpected results may be related to longer lead ages of pacing leads compared with defibrillator leads, as supported by our data (15.4 ± 8.5 years vs. 8.6 ± 4.4 years; p < 0.001). Looking more closely at our cohort of patients, we included only patients in the second and third quartiles of lead age to minimize the confounding effect of lead age on the rate of acute TRI. The pacing lead and defibrillator lead groups then demonstrate no significant difference in lead age (10.5 ± 2.4 years vs. 9.6 ± 1.7 years; p = 0.091) or in the rate of acute TRI (11.1% vs. 8.8%; p = 0.710). Therefore, lead age, and not the type of extracted leads, seems more closely related to the risk of acute TR aggravation.
Laser-assisted TLE is controversially considered a risk factor for TLE-related acute TR aggravation. Franceschi et al. (6) showed that the use of laser sheath as opposed to manual traction alone was the most powerful predictor of post-TLE significant TR. However, laser sheath is usually reserved for difficult cases, possibly with a longer lead age and more extensive fibrotic adhesion between the lead and TV apparatus. However, lead age was not compared between the laser sheath and manual traction groups in their study. Our data showed that lead age was significantly longer in patients treated by laser sheath versus patients treated by manual traction without other tools (11.9 ± 7.4 years vs. 6.4 ± 5.9 years; p = 0.012). If performed in patients with similar lead ages, manual traction alone could be associated with more frequent post-procedural acute TRI. On the contrary, in a retrospective study by Rodriguez et al. (10), no significant TRI was observed by TEE in all their 53 patients after laser lead extraction. However, the low incidence may have resulted from their relatively short lead age (3.2 ± 3.5 years). Our results revealed that the rates of TV damage and acute TRI increased progressively with lead age despite the use of laser sheath (Figure 3).
Extraction of multiple leads and use of more than 1 extraction tool were also shown to be risk factors for acute TRI in other studies (6,8). In our data, there was a trend toward an increased risk of acute TR aggravation in patients with extraction of ≥3 RV leads or use of ≥2 extraction tools. All these factors, along with a longer lead age, may be associated with more dense fibrotic adhesion and a greater risk of TV injury during TLE. Indeed, evidence of damage to the TV apparatus was detected by our TEE data only in the acute TRI group (0%, 0 of 184 vs. 29%, 7 of 24; p < 0.001). Unlike other data, older age (≥75 years), female sex, or the presence of vegetation on TV and/or leads turned out to be unrelated to the development of acute TRI in our cohort (6–8,10).
According to our results, the risk of TR aggravation seems to increase with lead age even 10 years after device implantation. In addition, TLE-related TR aggravation could also lead to longer hospital stays, right-sided heart failure, and worse long-term prognosis (6,17–21). The strategic use of lead abandonment instead of earlier TLE increases implantation duration and likely raises the risk of TRI, a consequence that should be considered. Therefore, the alternative strategy of earlier TLE during cardiac device revision, replacement, or upgrade may be more attractive (22).
This study has the limitations inherent to a retrospective analysis of a prospectively enrolled cohort. All TLE procedures were performed using the same treatment protocol, and TEE data were acquired immediately before and after extraction to minimize heterogeneities in TR assessment. Long-term TV status or the impact on the patient’s overall outcome was not evaluated because this was a short-term evaluation. Although our study group is a large cohort with TLE-related acute TRI, our sample size may still be insufficient to explore the risk factors for rarer long-term clinical events such as TV repair or replacement surgical procedures and new onset right-sided heart failure. As for the mechanisms of acute TRI observed in 24 patients, 7 patients showed TEE data suggesting TV injuries. In 1 patient without evidence of TV injury, aggravation of left ventricular systolic function accompanied by elevation of pulmonary artery pressure was noticed after the extraction of biventricular pacing leads. However, in the remaining 16 patients, the causes of acute TRI remained unclear. Therefore, a multicenter prospective study with a larger sample size should be conducted.
During the TLE procedure, the development of the acute TRI was detected in a significant minority of patients by TEE and was associated with more frequent damage to TV apparatus and a longer post-extraction hospital stay. The incidence of acute TRI was significantly elevated with lead implantation duration, which was identified as an independent risk factor for the acute TRI. Therefore, the time-dependent risk of TLE-related TR aggravation should be considered and should affect the strategic timing of TLE.
COMPETENCY IN MEDICAL KNOWLEDGE: The risk of TV damage or TR aggravation is likely to increase with lead implantation duration. TLE-related TR aggravation was associated with a longer hospital stay and could lead to worse long-term prognosis. Lead abandonment versus TLE strategy during cardiac device revision, replacement, or upgrade may need to include the time-dependent risk of TRI.
TRANSLATIONAL OUTLOOK: Further prospective studies are needed to explore the long-term clinical implications of TLE-related acute TR aggravation.
Dr. Tarakji has been on the medical advisory boards of Medtronic and AliveCor. Dr. Wilkoff has been a consultant for Medtronic, Abbott, and Philips. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- confidence interval
- cardiovascular implantable electronic devices
- odds ratio
- right ventricular
- transesophageal echocardiography
- transvenous lead extraction
- tricuspid regurgitation
- tricuspid regurgitation increase
- tricuspid valve
- Received May 22, 2018.
- Revision received July 16, 2018.
- Accepted July 17, 2018.
- 2018 American College of Cardiology Foundation
- Wazni O.,
- Wilkoff B.L.
- Wazni O.,
- Epstein L.M.,
- Carrillo R.G.,
- et al.
- Kim D.,
- Baek Y.S.,
- Lee M.,
- et al.
- Franceschi F.,
- Thuny F.,
- Giorgi R.,
- et al.
- Coffey J.O.,
- Sager S.J.,
- Gangireddy S.,
- Levine A.,
- Viles-Gonzalez J.F.,
- Fischer A.
- Givon A.,
- Vedernikova N.,
- Luria D.,
- et al.
- Rodriguez Y.,
- Mesa J.,
- Arguelles E.,
- Carrillo R.G.
- Kusumoto F.M.,
- Schoenfeld M.H.,
- Wilkoff B.L.,
- et al.
- Wilkoff B.L.,
- Kennergren C.,
- Love C.J.,
- Kutalek S.P.,
- Epstein L.M.,
- Carrillo R.
- Wazni O.,
- Tarakji K.G.,
- Wilkoff B.L.
- Lin G.,
- Nishimura R.A.,
- Connolly H.M.,
- Dearani J.A.,
- Sundt T.M. 3rd.,
- Hayes D.L.
- Hai T.,
- Lerner A.B.,
- Khamooshian A.
- Barakat A.F.,
- Wazni O.M.,
- Tarakji K.,
- et al.