Device Pocket Scar Predicts Transvenous Lead Extraction Difficulty

DOI: 10.19102/icrm.2015.061101

MELANIE MAYTIN, MD, MSc, FHRS, ROY M. JOHN, MD, PhD, FHRS and LAURENCE M. EPSTEIN, MD

Brigham and Women's Hospital, Cardiovascular Division, Boston, MA

Download PDF

Follow Us >> Tweet

ABSTRACT. The challenges and risks of transvenous lead extraction (TLE) of cardiovascular implantable electronic devices (CIEDs) are principally related to the body's foreign body response to endovascular leads. Despite this understanding, predictors of severe endovascular scar formation have not been clearly identified. The aim was to evaluate if the severity of the pocket scar may help predict endovascular scar and TLE difficulty. We performed a prospective analysis of consecutive patients undergoing TLE. Patient and procedural characteristics, classification of pocket scar severity, operator assessment of TLE difficulty, extraction time, and number of extraction sheaths (ESs) used are reported. Logistic and linear regression analyses were utilized to test the adjusted association between pocket scar and the combined endpoint and subjective difficulty assessed by the operator, respectively. Between November 2010 and February 2012,144 patients underwent TLE with assessment of pocket scar. The cohort was 63% male with mean age 62 ± 16 years. Average implant duration was 84 ± 53 months. Indications for TLE included infection 35%, malfunction 30%, upgrade 8%, and other 27%. Each incremental increase in pocket scar severity was associated with a twofold increase in TLE difficulty (OR 2.03; 95% CI 1.005—4.110). TLE difficulty endpoint correlated significantly with the operator's assessment of procedural difficulty (r = 0.74, p < 0.001). Operator scoring of TLE difficulty (1-10) was highly correlated with severity of CIED pocket scar (p = 0.001). The severity of the scar in the device pocket correlates with multiple ES use and long extraction times and operator quantification of extraction difficulty. CIED pocket scar may help predict TLE difficulty.

KEYWORDS. implantable cardioverter-defibrillator, lead extraction, lead management, outcomes, pacemaker.

Dr. Maytin has received research grants from Boston Scientific; participates in industry-sponsored research with Biotronik, Boston Scientific, and Medtronic; and is a consultant for Biotronik, Medtronic, Spectranetics, and St. Jude Medical. Dr. John participates in industry-sponsored research with Biosense Webster, Medtronic and Thermedical Inc and is a consultant for St. Jude Medical.
Dr. Epstein has received research grants from and is a consultant for Boston Scientific, Medtronic, Spectranetics and St. Jude Medical; and has equity in and served as a board member for Carrot Medical.
Manuscript received November 2, 2015, Final version accepted November 18, 2015.
Address correspondence to: Melanie Maytin, MD, Brigham & Women's Hospital, 75 Francis Street, Boston, MA 02115. E-mail: mmaytin@partners.org

Introduction

The demand for transvenous lead extraction (TLE) continues to rise; it is estimated to have reached an annual extraction rate of nearly 24,000 patients worldwide (B. Safyan, personal communication, Spectranetics). This growth is a consequence of expanding cardiovascular electronic implantable device (CIED) indications, an increase in CIED-related infections,^1,2 lead-related advisories and utilization, as well as newer tools and techniques with associated higher success rates. Despite these technologic advances, TLE can be associated with significant morbidity and mortality. Reported major complication and mortality rates with TLE vary widely across studies, ranging from 0.4% to 7.3%.^3–8

The difficulty and risk of TLE of CIED is predominantly related to the body's foreign body response to the endovascular leads. The response begins immediately after implantation with thrombus development along the lead(s). The foreign body response continues with fibrosis of the thrombus occurring next. Near-complete encapsulation of the leads with a fibrin sheath develops within 4-5 days of implant.^9–11 Robust fibrosis develops in areas of direct contact between the lead and the vasculature and endocardium. The most common adhesion sites include the venous entry site, the superior vena cava (SVC) and the electrode-myocardial interface, with multiple areas of scar tissue found in the majority of patients.¹² Calcification of the fibrotic lesions can occur with time, further cementing the adhesion site and increasing the difficulties and risks of the extraction.

Despite our understanding of the histopathologic changes following lead implantation, predictors of severe endovascular scar formation have not been clearly identified. We hypothesize that the severity of the pocket scar may help predict endovascular scar and, hence, TLE difficulty.

Methods

We identified a cohort of consecutive patients undergoing TLE at our institution between November 2010 and February 2012 who had the following analyzed prospectively: patient and procedural characteristics, classification of pocket scar severity (scale: mild, moderate, severe), operator subjective assessment of TLE difficulty (1–10; 1: least difficult, 10: extremely difficult), extraction time, and number of extraction sheaths (ESs) used. The lead extraction technique employed was the decision of the operator. All operators were highly skilled and well versed in all extraction modalities with a large volume of experience (>75 lead extractions/year). Logistic and linear regression analyses were utilized to test the adjusted association between pocket scar and the combined endpoint and subjective difficulty assessed by the operator, respectively. The study was approved by the Institutional Review Board of Brigham and Women's Hospital in Boston, MA.

TLE difficulty was defined as the combined endpoint of the need for two or more sheaths, and/or extraction time > 75th percentile. Classification of pocket scar severity was performed immediately upon entering the device pocket and prior to any attempts at lead extraction. Scar severity was defined subjectively as mild, moderate, or severe on the basis of the operator's assessment of scar thickness and quantity of scar as well as difficulty of lead dissection. The presence of a calcified scar was automatically classified as severe. Tissue from the pocket was sent to pathology for routine investigation in the majority of cases.

Outcomes were based upon the most recent Heart Rhythm Society lead management consensus¹³ and defined as follows: 1) complete procedural success if all targeted leads and lead material were removed from the vascular space; 2) clinical success if all targeted leads and lead material were removed but with retention of a small portion of the lead that does not negatively impact on outcome goals; and 3) failure if neither complete procedural nor clinical success could be achieved. Major complications were defined as death; cardiac or vascular avulsion or tear requiring thoracotomy, pericardiocentesis, chest tube, or surgical repair; pulmonary embolism quiring surgical intervention; respiratory arrest or anesthesia complication leading to prolongation of hospitalization; stroke; and pacing system-related infection of a previously non-infected site. Patients were followed in-hospital and procedural outcomes are reported.

Means, medians, or proportions for baseline clinical variables were calculated for the entire cohort. Continuous variables were expressed as the mean ± SD or median and interquartile ranges (IQRs). Fisher exact tests were used to compare categorical variables. Student t-tests were used to compare normally distributed continuous variables, and Wilcoxon rank sum tests were used for continuous variables that were not normally distributed. Logistic regression analysis was utilized to identify clinical variables associated with ES use. Variables in the multivariable logistic regression models included pocket scar severity, lead implant duration, number of leads extracted, indication for TLE, age at extraction, and gender. Age at extraction and implant duration were modeled as continuous variables after the linearity assumption was tested. All tests of significance were two sided, and p<0.05 was considered significant. Statistical data was performed using SAS version 9.2 (SAS Institute, Cary, NC).

Results

Between November 2010 and February 2012,126 patients underwent TLE with assessment of pocket scar. The cohort was 63% male with mean age 62 ± 16 years. Median implant duration was 72.2 (range 53.7-104.1) months. One-third of the TLE procedures were performed for infectious indications (35%). Indications for lead extraction included lead malfunction (30%), device system upgrade (8%), and other (27%) indications (including SVC syndrome, patient preference, and advisory leads). A total of 256 leads were removed in all. Pocket scar was classified as mild in the majority of patients (46.8%, see Figure 1a,b).

The baseline characteristics associated with pocket scar on unadjusted analysis included diabetes mellitus, indication for TLE, and incisional scar (Table 1). Diabetes mellitus was associated with more severe pocket scar (9% versus 56%, mild versus severe, p = 0.0003). Similarly, an association between indication for extraction and pocket scar severity was observed. The pocket scar was more severe in extraction cases performed for infection (both systemic and local) and less severe in extraction for lead malfunction or device system upgrade (p = 0.03). An incisional scar was significantly associated with and predictive of the severity of the pocket scar (p< 0.0001).

ES assistance was employed in 79% of cases. The average procedural time was 4.1 ± 9.1 min with an average 1.8 ± 0.9 leads removed/procedure. Many procedural characteristics and outcomes were associated with the severity of pocket scar (Table 2). The severity of pocket scar increased significantly with lead implant duration (p = 0.03). The number of endovascular leads and leads extracted were both positively associated with pocket scar severity. Similarly, the need for ES assistance, total number of sheaths used and TLE time were also positively associated with pocket scar severity. The difficulty of the extraction procedure was significantly associated with increasing pocket severity. While there was no significant difference between major complications and pocket scar severity, the only major complication (right hemothorax due to SVC tear requiring emergent sternotomy) occurred in the group with severe pocket scar.

Pocket scar severity, the number of leads extracted, and lead implant duration were significantly associated with TLE difficulty after adjusting for age, gender, and TLE indication (Table 3). Each incremental increase in pocket scar severity was associated with a twofold increase in TLE difficulty (OR 2.03; 95% CI 1.005, 4.110, p = 0.048). TLE difficulty endpoint correlated significantly with the operator's assessment of procedural difficulty (r = 0.74, p<0.001). Operator scoring of TLE difficulty (1-10) was highly correlated with severity of CIED pocket scar (p = 0.001).

Tissue was sent for pathology in 104 cases. The mean measured scar thickness was 0.4 ± 0.2 cm (range, 0.11.1 cm). The majority of pathology specimens had more than two findings (50%). The most common pathologic findings included fibrous, fibroconnective, and fibroadipose tissue (88%). Additional commonly identified pathologic findings included the following: inflammation, either chronic or acute (75%); fibrosis or scar (54%); foreign body reaction with either giant cells or granuloma formation (37%); hyalinization (11%); and calcification (11%). While operators were not able to grossly detect calcific changes, the presence of calcium on pathologic investigation was significantly associated with a higher number of extraction tools (1.7 ±0.9 versus 1.0 ±0.8, p = 0.03) and greater subjective operator assessment of extraction difficulty (6.6 ± 2.6 versus 3.5 ± 2.1, p = 0.008) with a trend towards longer extraction times (7.82 ± 10.53 min versus 4.3 ± 10.6 min, p = 0.21). The pathology specimen from the patient with the only major complication demonstrated calcification.

Discussion

We observed several clinical associations between more severe pocket scar and pre-procedural variables, notably diabetes mellitus and local and/or systemic device-related infection. Similarly, many procedural characteristics and outcomes were associated with the severity of pocket scar. Lead implant duration, the number of endovascular leads present and extracted, the need for ES assistance, the total number of sheaths used, TLE time, and the difficulty of the extraction procedure was significantly positively associated with pocket scar severity. On multivariate analysis, each incremental increase in pocket scar severity was associated with a twofold increase in TLE difficulty (OR 2.03; 95% CI 1.005-4.110, p = 0.048). Calcification was identified in pathologic specimens 11% of the time and was associated with more difficult extraction (higher number of extraction tools and greater operator assessment of extraction difficulty). Subjective notation of pocket scar calcification was performed by the operators, but this was not a prespecified endpoint and did not correlate with pathologic observation of calcification, likely due to sample size limitations. While only one major complication occurred, it occurred in a patient with severe pocket scar and calcification.

These findings are the first to demonstrate that pocket scar severity is a predictor of extraction difficulty on multivariate analysis. Moreover, this is the first study to demonstrate a positive association between pocket calcification and TLE difficulty. Biefer and colleagues¹⁴ developed a classification system for pocket adhesions and examined the correlation between pocket adhesions and TLE outcomes, specifically the number of extraction tools used and complete procedural success. In this retrospective analysis, the authors classified lead adhesions within the pocket on a scale of 0 to 3 and observed strong correlations with increasing pocket score and implant duration and number of extraction tools needed albeit with unadjusted results. This prospective analysis confirms the hypothesis generated by Biefer et al.,¹⁴ demonstrating pocket scar severity to be a predictor of TLE difficulty after controlling for age, gender, number of leads extracted, implant duration, and indication for TLE. While the pathologic changes associated with endovascular leads have been described previously,^9–11 this is first description of pathologic changes within the device pocket. Furthermore, this study is the first to demonstrate an association between calcific changes in the pocket and TLE difficulty. An association between younger age and calcified scar tissue has been described previously¹⁵ but we did not observe a similar association (age at implant: 50 ± 11 versus 55 ± 17 years, p = NS; age at TLE: 59 ± 11 versus 61 ± 16 years, p = NS).

Predictors of TLE difficulty have been evaluated in several prior single and multicenter studies.^5,16–18 Many variables previously identified as predictors of TLE difficulty were confirmed in this study, including implant duration^3,5,16–18 and the number of leads extracted.¹³ In contrast to prior studies, both local and systemic infection were associated with more severe pocket scar and increased TLE difficulty.⁵ The observed association between more severe pocket scar and infection is plausible given the pathologic changes that occur with infection including inflammation and granulation. Although this study was underpowered to detect relationship between major complications and pocket scar severity due to the small number of events, it is worth noting that the only major complication occurred in a patient with severe pocket severity and calcification within the pocket. Endovascular scar formation increases the difficulty and risk of TLE. We demonstrated a significant correlation between the severity of pocket scar, endovascular scar, and TLE difficulty. The identification of pocket scar severity and tissue calcification can help operators anticipate the challenges associated with a particular case and potentially change practice, preparation, and outcomes. The ability to predict TLE difficult at the onset of the case, i.e. at the time of pocket and lead dissection, may enable the operator to repeat a risk-benefit ratio assessment and potentially prevent adverse outcomes.

Conclusions

The severity of scar in the device pocket correlates with multiple ES use, long extraction times, and operator quantification of extraction difficulty. Similarly, pathologic changes in the pocket tissue, specifically calcification, are associated with increased TLE difficulty. The CIED pocket scar can help predict TLE difficulty.

Limitations

The cohort was limited to a single, high-volume, academic center and the experience may differ at other types of institutions. Although performed prospectively prior to attempts at lead extraction, the assessment of pocket scar severity was subjective and performed by a limited number of operators. While operator assessment of pocket scar severity was significantly correlated with objective measures of TLE difficulty (number of tools used and procedure time), this subjective variable may not be universally applicable to all operators of varying degrees of experience. Major complications were infrequent in this small series and conclusions cannot be drawn between pocket scar severity and complications.

References

1. 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(4):414–419. [CrossRef] [PubMed]
2. Cabell CH, Heidenreich PA, Chu VH, et al. Increasing rates of cardiac device infections among Medicare beneficiaries: 1990-1999. Am Heart J. 2004;147(4):582–586. [CrossRef] [PubMed]
3. 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(6):579–586. [CrossRef] [PubMed]
4. Bracke FA, Meijer A, van Gelder LM. Lead extraction for device related infections: A single-centre experience. Europace. 2004;6(3):243–247. [CrossRef] [PubMed]
5. 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(9):1348–1357. [CrossRef] [PubMed]
6. Jones SO, Eckart RE, Albert CM, Epstein LM. Large, single-center, single-operator experience with transvenous lead extraction: Outcomes and changing indications. Heart Rhythm. 2008;5(4):520–525. [CrossRef] [PubMed]
7. Kennergren C, Bjurman C, Wiklund R, Gabel J. A single-centre experience of over one thousand lead extractions. Europace. 2009;11(5):612–617. [CrossRef] [PubMed]
8. Bongiorni MG, Soldati E, Zucchelli G, et al. Transvenous removal of pacing and implantable cardiac defibrillating leads using single sheath mechanical dilatation and multiple venous approaches: High success rate and safety in more than 2000 leads. Eur Heart J. 2008;29(23):2886–2893. [CrossRef] [PubMed]
9. Huang TY, Baba N. Cardiac pathology of transvenous pacemakers. Am Heart J. 1972;83(4):469–474. [CrossRef] [PubMed]
10. Robboy SJ, Harthorne JW, Leinbach RC, Sanders CA, Austen WG. Autopsy findings with permanent pervenous pacemakers. Circulation. 1969;39(4):495–501. [CrossRef] [PubMed]
11. Esposito M, Kennergren C, Holmstrom N, Nilsson S, Eckerdal J, Thomsen P. Morphologic and immunohistochemical observations of tissues surrounding retrieved transvenous pacemaker leads. J Biomed Mater Res. 2002;63(5):548–558. [CrossRef] [PubMed]
12. Smith HJ, Fearnot NE, Byrd CL, Wilkoff BL, Love CJ, Sellers TD. Five-years experience with intravascular lead extraction. U.S. Lead Extraction Database. Pacing and clinical electrophysiology. Pacing Clin Electrophysiol. 1994;17(11 Pt 2): 2016–2020. [CrossRef] [PubMed]
13. 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(7):1085-1104. . [CrossRef] [PubMed]
14. Biefer HR, Hurlimann D, Grunenfelder J, et al. Generator pocket adhesions of cardiac leads: Classification and correlation with transvenous lead extraction results. Pacing Clin Electrophysiol. 2013;36(9):1111–1116. [CrossRef] [PubMed]
15. Smith MC, Love CJ. Extraction of transvenous pacing and ICD leads. Pacing Clin Electrophysiol. 2008;31(6):736–752. [CrossRef] [PubMed]
16. Maytin M, Love CJ, Fischer A, et al. Multicenter experience with extraction of the Sprint Fidelis implantable cardioverter-defibrillator lead. J Am Coll Cardiol. 2010;56(8): 646–650. [CrossRef] [PubMed]
17. Maytin M, Wilkoff BL, Brunner M, et al. Multicenter Experience with Extraction of the Riata/Riata ST ICD Lead. Heart Rhythm. 2014;11(9):1613–1618. [CrossRef] [PubMed]
18. Segreti L, Di Cori A, Soldati E, et al. Major predictors of fibrous adherences in transvenous implantable cardioverter-defibrillator lead extraction. Heart Rhythm. 2014;11(12): 2196–2201. [CrossRef] [PubMed]