Safety and feasibility of same‐day discharge following uncomplicated transvenous lead extraction

Transvenous lead extraction (TLE), while mostly a safe procedure, has risk of serious periprocedural complications. As such, overnight hospitalization remains a routine practice. In our center, we routinely discharge patients on the same day following an uncomplicated TLE.


| INTRODUCTION
Transvenous lead extraction (TLE) is an important strategy in management of patients with cardiac implantable electronic devices. 1 TLE is indicated in patients with an infected device system, either systemic or local pocket infection.TLE is also often indicated and performed for patients with lead malfunction or dislodgement, superfluous leads, noninfective lead-related complications (i.e., vein occlusion/stenosis/thrombosis, chronic pain, phrenic nerve capture), or those requiring a device upgrade.A TLE procedure may simply require gentle traction on the leads or may require more advanced tools such as powered sheaths (i.e., laser, radiofrequency, or mechanical rotational) to free the lead from the fibrous tissue or calcifications, which can be considerable for older leads and increase the risk of the TLE procedure.In addition, the risk of TLE may be elevated based on patient factors such as age, body mass index, and co-morbid conditions.Due to the potential for an elevated risk, after TLE procedures patients are commonly observed overnight before discharge.
Alternatively, same-day discharge (SDD) might be considered, having the potential to reduce the costs associated with the procedure and improve patient satisfaction. 24][5] In our center, we routinely discharge patients the same day after an uncomplicated TLE: herein, we sought to review our experience with SDD in selected patients following uncomplicated TLE, to shed more light on patient selection and to evaluate the safety of our practice.

| METHODS
This is a single-center, retrospective, cohort study of consecutive patients undergoing uncomplicated TLE at St. David's Medical Center in Austin, Texas between January 2019 and December 2021 (Figure 1).Patients with intraprocedural and immediate (within 2 h following the procedure end time) complications were excluded from the study.SDD was considered to have occurred if patients were discharged on the same day of the TLE procedure.SDD patients were compared with those who were not discharged on the same day of the lead extraction procedure (non-SDD).
Baseline demographics, as well as clinical characteristics, procedural details, and in-hospital/postdischarge outcomes were collected by chart review as part of our clinical database of patients undergoing electrophysiology procedures, which is approved by the St. David's HealthCare Institutional Review Board (#18-05-07).TLE is usually performed from the pocket with a superior approach via the subclavian vein.After opening the pocket, leads are disconnected from the device and dissected free from scar tissue down to the venous entry point, where the anchor suture sleeves are removed.Using a stepwise approach (Figure 2), 6 simple manual traction, either by itself or over a standard stylet, is first attempted.If this is unsuccessful, the lead is cut and a locking stylet (Lead Locking Device -LLD -EZ; Philips) is advanced into the lead and secured to the insulation.If manual traction over a locking stylet is still unsuccessful, powered sheaths are employed.

| TLE procedure at our institution
Based on the operator preference, laser (GlideLight; Philips) and/ or mechanical rotational (TightRail, Philips; or Evolution, Cook Medical) sheaths are used to achieve complete TLE.A transfemoral approach is used for snaring (Needle's Eye Snare, Cook Medical; EN Snare, Merit Medical; or Amplatz GooseNeck Snare, GIANNI ET AL. | 279 Medtronic) either to retrieve cut or broken leads after a failed superior approach or to maintain vascular access in the setting of venous occlusion or severe stenosis. 7ter TLE is completed, device reimplantation is performed unless otherwise indicated (systemic infection, device no longer needed).Upon completion of the procedure, all femoral vascular access sheaths are removed and hemostasis is achieved with manual pressure or using vascular access closure devices (collagen plug-or suture-based).Patients are then transferred to the recovery area, where frequent noninvasive blood pressure and ECG monitoring are continued for at least 2-h after the procedure.
Patients are ambulated after 2 h of bed rest if a closure device was used or after 4-6 h if the sheaths are removed manually.At the time of ambulation, patients are either discharged or remain in the hospital for overnight observation, according to the operator's and patient's preference.

| Study endpoints
The main study endpoint was to compare the rate of nonimmediate (>2 h) major procedure-related complications between patients in the SDD group and those in the non-SDD group at 1-, 7-, 30-days, and overall.Procedure-related complications were defined as any adverse event definitely or possibly related to the TLE procedure, 1 including but not limited to procedure-related death or death of unknown cause, cerebrovascular accident, acute heart failure, pericardial effusion, pericarditis, tricuspid valve injury, pulmonary embolism, pneumothorax, hemothorax, surgical site infection, device infection, pocket hematoma, deep venous thrombosis, lead dislodgement, femoral vascular access-related complications (hematoma, arteriovenous fistula, pseudoaneurysm), and acute kidney injury.Major complications were defined as those associated with or rehospitalization after discharge from the index hospitalization), or that resulted in (ii) death or (iii) permanent sequelae.
Secondary endpoint was to identify clinical and procedural predictors of SDD in our patient population to better understand what ed the operator's decision to discharge the patient the same day.

| Statistical analysis
Continuous variables are presented as mean ± standard deviation or median (interquartile range), according to their distribution; categorical variables are expressed as count (percentage). 8Univariate analysis (unpaired t test and Fisher's exact test) was used to test for differences between the two groups.Multivariable logistic regression was used to identify independent predictors of SDD by evaluating the association of SDD with the variables found to be significant on univariate analysis.
All tests were two-tailed and conducted at an α level of .05.
Statistical analysis was performed with GraphPad Prism 9.5 for Mac (GraphPad Software).
Baseline characteristics are shown in Table 1.The mean age of patients was 67 ± 14 years, 112/264 (42%) patients were female.
The indication for TLE was more commonly noninfectious in patients of the SDD group when compared with those of the non-SDD group (93% vs. 46%; p value < .0001).
Procedural characteristics are shown in Table 2.The number of operators performing the procedure, type of implanted device, lead dwell time, LED score (number of extracted leads + lead dwell time in years +1 if dual-coil internal-cardioverter defibrillator [ICD] lead −1 if lead vegetation), 9 approach for TLE (superior, from the pocket vs. transfemoral approach), and use of powered sheaths were comparable between the two groups.Notably, in our cohort, a trans-femoral approach was used in 11 patients, mostly to maintain vascular access for subsequent same-procedure reimplant (4 patients in the SDD group; 4 patients in the nSDD group), and more rarely for lead retrieval/extraction (3 patients of the nSDD group, 2 retrieval of broken leads, 1 extraction of an abdominal device's infected lead 10 ).Patients in the SDD group had fewer leads to extract compared with those of the non-SDD group 23.4 ± 25.9 min, difference −11.3 min, 95% CI −17.1 to −5.4 min, respectively), albeit with a comparable number of extracted leads and laser pulses needed for laser-assisted TLE.Additionally, the procedure end time was more likely before 12:00 in patients in the SDD group when compared with those in the non-SDD group (116/141 or 82% vs. 17/35 or 49%, p value = .0001).On multivariate analysis, the only predictor of SDD was a procedural end time before 12:00 (OR 3.83; 95% CI 1.40-10.4).

| Safety outcomes
Safety outcomes are shown in Tables 3 and 4

| DISCUSSION
Our study shows that SDD is safe after an uncomplicated TLE procedure completed in the morning in patients without device infection.
2][13] SDD is appealing to both patients and hospitals, leading to increased patient satisfaction and cost efficiency.
However, before widespread adoption, the safety of this approach for a potentially high-risk procedure, such as TLE, must be established to prevent patient harm.
Two groups have recently published their experience SDD after TLE. 4,5In their case series, Atteya et al. described the characteristics and outcomes of 22 patients who were discharged the same day following TLE for a noninfectious indication, with no reported 30-day procedure-or device-related complications. 4 In their case-control study of patients undergoing elective TLE for a noninfectious indication, Dagher et al. compared those who were discharged the same day (n = 80) with those who stayed overnight due to the occurrence of immediate procedure-related complications (n = 31), confirming the safety of an SDD approach, with no reported major procedure-related complications. 5Our study adds to the existing body of knowledge, because it is the first one evaluating the safety of SDD in patients following uncomplicated TLE (i.e., patients with immediate complications were not included) for any indication (i.e., patients with an infectious indication were included).Moreover, it is a larger study with more balanced groups, producing more precise estimates of safety outcomes.
While catastrophic complications can occur in the setting TLE (most notably SVC and cardiac tear), these usually occur during the procedure and in the absence of immediate (<2 h) adverse events, a strategy of SDD is safe.This study shows that patients discharged the same day and those who were discharged after at least overnight observation had similar major procedure-related complication rates in the early (within 1 day) postprocedural period.The rate of 7-and 30-days major procedure-related complications was statistically significantly higher in the non-SDD Importantly, neither patient's comorbidities nor procedure complexity influenced the decision to discharge the same day after correcting for procedure end time and noninfectious indication.Of note, as a high-volume center with a strong connection with an established surgical program, TLE procedures were complex on average, as shown by a mean LED score >10. 14Nevertheless, the LED score was comparable between groups, as was the rate of patients who required use of powered sheaths, and the use of a trans-femoral approach (which was mostly employed to maintain vascular access in the setting of venous occlusion or severe stenosis).
These findings are confirmed when considering only elective procedures performed for a noninfectious indication, the same cohort investigated in the studies previously published on the Note: Notably, the total number of complications is higher than the number of patients with complications (see this table) as some patients had more than one complication.
Abbreviations: ICD, internal-cardioverter defibrillator; non-SDD, nonsame-day discharge; SDD, same-day discharge.a One for acute heart failure, one for femoral vascular access-related complication.
b One appropriate ICD shock, one non-ST elevation myocardial infarction.
T A B L E 4 Rate of patients with major procedure-related complications at 1-, 7-, and 30-days.| 285 feasibility of SDD for TLE. 4,5More specifically, in this cohort, we found that the only significant predictor of SDD remains a procedure end-time before 12:00.This does not confirm the findings of Dagher et al., which identified a higher procedure complexity (more extracted leads, with longer dwell times, using more advanced tools and techniques) as a predictor of overnight stay. 5However, while our study focuses on patients undergoing uncomplicated TLE, Dagher et al. also included patients with immediate procedure-related complications, which was the main reason for overnight stay in their cohort.This is a potential confounder that is not controlled by the authors in their analysis and can explain the association between a higher procedure complexity and overnight stay, both of which are associated with a higher rate of immediate complications.In our study, we control for this confounder by simply excluding those with immediate complications, patients who would not have been discharged the same day, irrespective of other clinical and procedural considerations.

| Study limitations
Our study is primarily limited by its single-center, retrospective, observational nature.Nevertheless, this is a large case-control study that provides supportive evidence on the safety of SDD in selected patients undergoing uncomplicated TLE by showing noninferior outcomes to patients who were not discharged on the same day.
Thus, our data might help guide future prospective studies evaluating the safety of a widespread adoption of this practice.Additionally, patients included in the study are consecutive and, except for absence of immediate complications, nonselected.As such, this cohort represents the general population of patients undergoing uncomplicated TLE, and the study's findings can be generalized to real-world settings.

| CONCLUSIONS
Following an uncomplicated TLE procedure performed in a center with an established surgical program, SDD in selected patients (i.e., those without device infection and when the TLE procedure is completed in the morning) is feasible and safe.
TLE procedures are performed in our center in a hybrid setting, with surgical backup and packed red cells available in the room to promptly intervene in case of emergent bleeding in the setting of superior vena cava (SVC) or cardiac tear.Six different electrophysiologists performed TLE during the study period, with most procedures performed by four operators (RCC, 179/265 or 68%; AAA, 71/265 or 27%; MAB 32/265 or 12%; PRC 30/265 or 11%).All procedures are performed under general anesthesia.In addition to routine monitoring (pulse oximetry, capnography, 12lead ECG), invasive arterial pressure monitoring (a 4-Fr or 5-Fr femoral arterial line) and continuous transesophageal echocardiography are used in all patients for early detection of pleural and pericardial bleeding.A large-bore (8-Fr) femoral venous access is also obtained and, along with the arterial access, is ready to be used for resuscitation and to quickly establish cardiopulmonary by-pass in case of major vascular or cardiac injury.If indicated, additional femoral venous access sites are obtained: (i) to quickly deploy a compliant balloon (Bridge Occlusion Balloon; Philips) in case of SVC tear, (ii) for temporary pacing in pacing-depended patients, and/or (iii) for transfemoral snaring if extraction from the pocket is not successful.
(i) hospitalization (prolonged index hospitalization before discharge F I G U R E 1 Study flowchart.TLE, transvenous lead extraction.

|
283       responsive to mechanical ventilation and a patient with concomitant left ventricular assist device infection), and nine rehospitalizations (three for wound dehiscence: two requiring surgical debridement and intravenous antibiotic treatment, and one managed medically with intravenous antibiotic treatment; two for pocket hematomas requiring evacuation; two for femoral vascular access-related complication: one patient with bacteremia who developed pseudoaneurysm successfully treated with thrombin injection, and one symptomatic artero-venous fistula; one for non-ST elevation myocardial infarction treated with percutaneous coronary intervention; and one for appropriate ICD shock).

F I G U R E 3
Clinical and procedural factors influencing the clinical-decision making to SDD.CI, confidence interval; COPD, chronic obstructive pulmonary disease; non-SDD, non-same-day discharge; SDD, same-day discharge.group, reflecting the sicker population of patients included in this cohort.Indeed, in this group, most complications within 7-days (9 out of 11) occurred in patients who underwent TLE for bacteremia/lead infection.Importantly, there were no deaths in the SDD group and readmissions for procedure-related complications were rare, occurring in 7/152 (4.6%) patients for late presenting complications which would not have deferred next-day discharge.More specifically, after SDD, there were no major procedure-related complications within 1 day and patients were rehospitalized for pocket hematoma or surgical site infection within 7 days and for pericarditis, lead dislodgement, or secondary device infection within 30 days.These are all late complications, that would not have been detected by keeping the patient in the hospital overnight.When looking at patient selection, predictors analysis shed a light on the operators' clinical-decision making process: the main factors influencing the decision to discharge the same day were a procedure performed in the morning (i.e., end time before 12:00) and for a noninfectious indication.The former factor is explained by postprocedural logistics, as a minimum of 2 h of observation is required after the procedure but might be longer in case of lingering anesthesia effects or need for prolonged bed rest if vascular closure devices are not used.The latter factor encompasses the cohort of patients with device infection, those likely to require hospitalization due to the systemic involvement of the infectious process and the need for intravenous antibiotic treatment and supportive care.

T A B L E 3
Major procedure-related complications in the study population.
. Over 30 days, there were a total of 31 major procedure-related complications occurring in 28 patients, 8 complications in 7 patients of the SDD group, and 23 complications in 21 patients of the non-SDD group.On a patient T A B L E 1 Baseline characteristics.Procedural characteristics.
Abbreviations: BMI, body mass index; CI, confidence interval; COPD, chronic obstructive pulmonary disease; LVEF, left ventricular ejection fraction; non-SDD, non-same-day discharge; SDD, same-day discharge; TLE, transvenous lead extraction.aBMI≥30kg/m 2 .bkidneyinjuryresulting in prolonged hospitalization in a patient with bacteremia.Between 2-and 7-days, three patients in the SDD group were rehospitalized (one for pocket hematoma requiring evacuation; one for wound dehiscence requiring closure; and one for concomitant deep venous thrombosis and pocket hematoma requiring evacuation).In the non-SDD group, eight patients suffered major procedure-related complications: two deaths (one patient with bacteremia who died for acute on chronic heart failure; one patient with bacteremia who died for hemorrhagic shock in the setting of severe coagulodislodgement requiring lead revision; and one for device infection and concomitant acute heart failure which required removal of the recently reimplanted system).In the non-SDD group, there were 11 major procedure-related complications: two death of unknown cause (two patients with bacteremia who were still hospitalized after TLE: one patient who developed acute on chronic respiratory failure notT A B L E 2Abbreviations: AK, air kerma; DAP, dose area product; LED, lead extraction difficulty; non-SDD, non-same-day discharge; SDD, same-day discharge.aLED score is defined as number of extracted leads + oldest lead dwell time in years +1 if dual-coil ICD lead −1 if lead vegetation.bComplete success is defined as complete removal of a lead; failure or incomplete removal (including with small remnant) are not considered complete success.c Radiation dose data are available in 207/265 (78%) patients, 125/153 (82%) of the SDD group, and 82/112 (73%) of the non-SDD group.GIANNI ET AL.