*Article* **The Value of Left Ventricular Mechanical Dyssynchrony and Scar Burden in the Combined Assessment of Factors Associated with Cardiac Resynchronization Therapy Response in Patients with CRT-D**

**Tariel A. Atabekov, Mikhail S. Khlynin \*, Anna I. Mishkina, Roman E. Batalov, Svetlana I. Sazonova, Sergey N. Krivolapov, Victor V. Saushkin, Yuliya V. Varlamova, Konstantin V. Zavadovsky and Sergey V. Popov**

> Cardiology Research Institute, Tomsk National Research Medical Center, Russian Academy of Sciences, Kievskaya St., 111a, Tomsk 634012, Russia

**\*** Correspondence: mskhlynin@mail.ru

**Abstract:** Background: Cardiac resynchronization therapy (CRT) improves the outcome in patients with heart failure (HF). However, approximately 30% of patients are nonresponsive to CRT. The aim of this study was to determine the role of the left ventricular (LV) mechanical dyssynchrony (MD) and scar burden as predictors of CRT response. Methods: In this study, we included 56 patients with HF and the left bundle-branch block with QRS duration ≥ 150 ms who underwent CRT-D implantation. In addition to a full examination, myocardial perfusion imaging and gated blood-pool single-photon emission computed tomography were performed. Patients were grouped based on the response to CRT assessed via echocardiography (decrease in LV end-systolic volume ≥15% or/and improvement in the LV ejection fraction ≥5%). Results: In total, 45 patients (80.3%) were responders and 11 (19.7%) were nonresponders to CRT. In multivariate logistic regression, LV anterior-wall standard deviation (adjusted odds ratio (OR) 1.5275; 95% confidence interval (CI) 1.1472–2.0340; *p* = 0.0037), summed rest score (OR 0.7299; 95% CI 0.5627–0.9469; *p* = 0.0178), and HF nonischemic etiology (OR 20.1425; 95% CI 1.2719–318.9961; *p* = 0.0331) were the independent predictors of CRT response. Conclusion: Scar burden and MD assessed using cardiac scintigraphy are associated with response to CRT.

**Keywords:** heart failure; left bundle-branch block; mechanical dyssynchrony; response to cardiac resynchronization therapy; gated SPECT myocardial perfusion imaging

#### **1. Introduction**

Heart failure (HF) is a rapidly growing public health issue, with an estimated prevalence of more than 37.7 million individuals globally [1]. In the developed world, this disease affects approximately 2.0% of the adult population [2]. In the United States, the total percentage of the population with HF is projected to rise from 2.4% in 2012 to 3.0% in 2030 [3,4]. In Russian Federation, the prevalence of chronic HF (CHF) is 10.2% [5]. The main cause of CHF is coronary heart disease, which accounts for about 70.0%, and the remaining 30.0% are nonischemic heart diseases [6]. More than two decades of research has established the role of cardiac resynchronization therapy (CRT) in medically refractory, mild to severe systolic HF with abnormal QRS duration and morphology [7]. The prolongation of QRS (120 ms or more) occurs in 14.0% to 47.0% of HF patients and the ventricular conduction disturbance, most commonly the left bundle-branch block (LBBB), is present in approximately one-third of HF, leading to the mechanical dyssynchrony (MD) of ventricles [6,8]. Prospective randomized studies of patients with both ischemic HF (IHF) and nonischemic HF (NIHF) have shown that CRT translates into long-term clinical benefits, such as improved quality of life, increased functional capacity, reduction in hospitalization for HF, and overall mortality [9–11]. These patients qualified as responders to

**Citation:** Atabekov, T.A.; Khlynin, M.S.; Mishkina, A.I.; Batalov, R.E.; Sazonova, S.I.; Krivolapov, S.N.; Saushkin, V.V.; Varlamova, Y.V.; Zavadovsky, K.V.; Popov, S.V. The Value of Left Ventricular Mechanical Dyssynchrony and Scar Burden in the Combined Assessment of Factors Associated with Cardiac Resynchronization Therapy Response in Patients with CRT-D. *J. Clin. Med.* **2023**, *12*, 2120. https:// doi.org/10.3390/jcm12062120

Academic Editor: Andrea Di Cori

Received: 6 February 2023 Revised: 1 March 2023 Accepted: 7 March 2023 Published: 8 March 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

CRT [12,13]. However, CRT is effective in 70.0% of patients, and the remaining 30.0% do not respond to the device therapy [7,14]. This finding suggests that the existing criteria for patient selection to CRT are not always effective. In this regard, it is necessary to search for additional selection criteria or prognostic markers for CRT response in patients with CHF.

Some studies have shown that the ventricular MD assessment using transthoracic echocardiography (TTE), magnetic resonance imaging (MRI), computed tomography (CT), and single-photon emission computed tomography (SPECT) can play a key role in the prediction of CRT response [15–19]. However, according to other studies, TTE indicators are not reliable CRT response predictors [20]. The application of MRI, despite its high accuracy and information content, is limited by its high cost and complexity of cardiac protocols, and it is an operator-dependent method [21]. By contrast, gated myocardial perfusion imaging (MPI) and gated blood-pool SPECT (gBPS) are simpler, have higher reproducibility, and allow for the detection of the area of ventricular MD and impaired myocardial flow and scarring [22]. The predictive value of MD and scar burden assessed via MPI remains disputable. In some studies, MD is correlated with the positive response to CRT in patients with HF of ischemic and nonischemic etiologies [23,24]. However, another work shows the absence of MD prognostic significance in patients with IHF [25,26]. The good prognostic value of scar burden assessed via cardiac scintigraphy has been demonstrated in several studies [27]. However, papers devoted to the combined assessment of the factors associated with reverse remodeling, which can be used for the improved selection of patients for CRT, are rare [28].

The aim of this study was to identify the significance of MD and scar burden assessment using MPI and gBPS as a combined predictor of CRT response in patients with HF of ischemic and nonischemic etiologies.

#### **2. Materials and Methods**

#### *2.1. Patient Population and Study Design*

Patients with indications for CRT according to the ESC guidelines were included in this clinical, nonrandomized, open, prospective study [29]. The inclusion and exclusion criteria were determined in accordance with the research project "Single-Photon Emission Computed Tomography for Prediction and Evaluation of Cardiac Resynchronization Therapy Efficacy in Chronic Heart Failure Patients" (ClinicalTrials.gov, NCT03667989). Patients who were included in the study met the following criteria: the presence of HF (ischemic or nonischemic etiology), sinus rhythm, permanent LBBB with QRS duration ≥150 ms, New York Heart Association (NYHA) functional class (FC) of HF II-III, left ventricular ejection fraction (LVEF) ≤35%, and optimal medical therapy for at least 3 months. Individuals with NYHA FC of HF 0, I, IV, decompensated HF, recent myocardial infarction (less than 3 months), recent revascularization (within last 3 months), other cardiac intervention, acute HF decompensation, right bundle-branch block, previously implanted pacemaker or cardioverter–defibrillator, severe comorbidity, cognitive impairment, or with indications of revascularization and heart transplantation, as well as patients under 18 years of age, were excluded. HF etiology was considered ischemic in the presence of significant coronary artery disease (≥50% stenosis in one or more of the major coronary arteries) and/or a history of myocardial infarction or prior revascularization.

All patients underwent full physical examination (6 min walk distance test (6MWDT), electrocardiography (ECG), transthoracic echocardiography (TTE), Holter ECG monitoring, coronary angiography, and blood analyses), MPI with 99mTc–methoxy isobutyl isonitrile (99mTc–MIBI) and gBPS before device implantation. In all cases, CRT devices with the defibrillation function (CRT-D) were implanted, according to the ESC guidelines [29]. All patients received the basic therapy in accordance with the present guidelines. The follow-up was performed 6 months after CRT-D implantation.

#### *2.2. Consent*

The study was carried out in accordance with the principles of the Helsinki Declaration and with the standards of good clinical practice. The study protocol was approved by the local ethics committee. All the participants received written informed consent prior to the study inclusion. Ethical approval by the hospital review committee and patient consent according to the institutional guidelines were obtained.

#### *2.3. Minute Walk Distance Test*

The assessment of HF FC was performed in accordance with the NYHA criteria, using 6MWDT before and 6 months after CRT-D implantation. For the analysis, the walking distance in meters and NYHA FC were used, and the following ranking was used:


#### *2.4. TTE Acquisition and Analysis*

TTE with the intracardiac hemodynamic parameter assessment was performed using Philips HD15 PureWave (the Netherlands) ultrasound machine before and after 6 months of CRT-D implantation. The examination was carried out from standard positions with the determination of the left atrium (LA) and right ventricle (RV) sizes, interventricular septum (IVS) and LV posterior-wall (LVPW) thickness, LV end-systolic dimension (LVESD), LV enddiastolic dimension (LVEDD), LV end-systolic volume (LVESV), LV end-diastolic volume (LVEDV), the index of myocardial mass (IMM), LVEF, right ventricular systolic pressure (RVSP), stroke volume (SV), LV end-systolic index (LVESI), LV end-diastolic index (LVEDI), the left atrial index (LAI), and the right atrial index (RAI). The mitral, tricuspid, and aortic valve functions, as well as the right and left ventricular contractility, were assessed.

#### *2.5. Scintigraphic Data Acquisition*

The scintigraphy examination was performed using CZT SPECT/CT (GE Discovery 570C, GE Healthcare, Haifa, Israel) with low-energy multi-pinhole collimators and 19 stationary detectors [30]. Each detector contained 32 × 32 pixelated (2.46 × 2.46 mm) CZT elements. The energy window was symmetrically centered to ± 20% of the 140 keV Tc-99m photopeak. The images were reconstructed on the dedicated workstation (Xeleris 4.0; GE Healthcare, Haifa, Israel).

The time interval between MPI and gBPS examinations ranged from 1 to 2 days. MPI was performed first in all patients.

#### *2.6. MPI Acquisition and Analysis*

MPI was performed according to the standard ECG-gated (16 frames/cardiac cycle) rest protocol [30] approximately 60 min after the administration of 138–357 MBq of 99mTc– MIBI. All patients were imaged in a supine position with arms placed over their heads for an acquisition time of 7 min. A low-dose CT scan (120 kV, 20 mA) was performed for the attenuation correction. All images were reconstructed using the iterative reconstruction (60 iterations; Green OSL α 0.7; Green OSL β 0.3) and Butterworth post-processing filter (frequency 0.37; order 7) in a 70 × 70-pixel matrix with 50 slices. MPI data were processed using Corridor 4DM (University of Michigan, Ann Arbor, MI, USA) software. The reconstructed MPI included the standard cardiac short and vertical and horizontal long axes and the 17-segment bull's eye map. Each of the 17 segments was scored based on a semiquantitative 5-point scoring system (from 0—normal uptake to 4—no radiotracer uptake) [31]. The summed rest score (SRS) of all segments was quantified. Scar burden was defined as all segments with abnormal radiotracer uptake at rest.

#### *2.7. gBPS Acquisition and Analysis*

gBPS was performed following in vivo labeling patient's red blood cells with a 99mTc–pertechnetate dose of 555–720 MBq [32]. The data were acquired with ECG gating (16 frames/cardiac cycle). The patients were imaged in a supine position with arms placed over their heads for an acquisition time of 10 min. No attenuation correction was used. Images were reconstructed using iterative reconstruction (60 iterations; Green OSL α 0.7; Green OSL β 0.3) and a Butterworth post-processing filter (frequency 0.52; order 5) in a 70 × 70-pixel matrix with 57 slices. The image quantification and phase analysis were obtained with the Quantitative Blood-Pool SPECT 2009.0 (Cedars-Sinai Medical Center, Los Angeles, CA, USA) software, which allowed for an evaluation of the functional variables of LV and RV. Ventricular contours were adjusted manually when required. The following parameters for both ventricles were determined: the peak emptying rate (PER, expressed as EDV/s), the peak filling rate (PFR, EDV/s), and the second peak filling rate (PFR2, EDV/s).

The severity of intra- and interventricular dyssynchrony was evaluated using the Fourier transform method. Global MD indices were evaluated for ventricles' phase standard deviation (SD), histogram bandwidth (HBW), and phase entropy (PE). Interventricular dyssynchrony (IVD) was calculated according to the histogram peak of LV and RV. Moreover, the following regional MD indices were assessed: SD and E. The regional analysis of LV MD was based on the assessment of the free wall (FW), anterior wall (AW), lateral wall (LW), inferior wall (IW), and septal wall (SW) of LV. The mean effective radiation dose for the entire study protocol was 7.48 ±1 mSv (range 5.1–10.3 mSv) per patient.

#### *2.8. CRT-D Implantation and Programming*

The active-fixation atrial (AL) and defibrillation leads (DLs), as well as the passivefixation LV lead, were positioned under fluoroscopic guidance using a transvenous approach. DL was implanted into the right ventricular apex or interventricular septum. Lead positions were confirmed via fluoroscopy in the posteroanterior (PA) and left anterior oblique (LAO) view and the intraoperative threshold testing. The capture threshold, sensing amplitude, and impedance measurements of the leads were performed using the pacing system analyzer (Medtronic, Minneapolis, MN, USA) with sterile crocodile clip cables.

The implantation of the LV pacing lead was performed by cannulating one of the tributaries of the coronary sinus using the delivery system. The venogram was performed in AP and LAO projections. Access to the target branch within the coronary sinus and subsequent advancement of the LV lead was performed using an "over-the-wire" technique. In case of troubles with the target vein branch cannulation, we used the "interventional" technique with inner catheters (subselector). The intraoperative threshold testing was performed for checking the optimal pacing threshold and phrenic nerve stimulation after LV lead was positioned to the target vein.

CRT-D programming was carried out in accordance with international standards [33]. In every CRT-D device, the monitoring zone was programmed with a heart rate of 140–170 beats per minute (bpm) for more than 50 consecutive cycles without antitachycardia pacing (ATP) and shock therapy. The ventricular tachycardia (VT) zone was programmed for 170–200 bpm with 30 cycles and with ATP (≥1 burst pacing and ≥1 ramp pacing) and shock therapy (first shock with the submaximal shock discharge). The ventricular fibrillation (VF) zone was programmed for ≥201 bpm with 12 cycles and with ATP during CRT-D charging and the maximum shock discharge. The atrioventricular (AV) delay interval was programmed 20–40 ms less than the native AV delay. Biventricular pacing was switched on in all patients, and the V–V interval (left ventricle → right ventricle) was optimized using the surface ECG to obtain the narrowest QRS complex.

#### *2.9. Clinical Follow-Up, CRT-D Data Acquisition, and Analysis*

The follow-up information was acquired for 56 of the 64 patients who were included in the prospective analysis. The information about arrhythmic events was determined from CRT-D interrogation reports. The arrhythmic events (VT, VF, atrial fibrillation, and the appropriate and inappropriate ICD therapy) and CRT-D lead system parameters were evaluated at follow-up. The flowchart of the study is shown in Figure 1. The primary endpoint was a decrease in LVESV ≥15% or/and an improvement in LVEF ≥5% [26,34].

**Figure 1.** Study flowchart: HF—heart failure; MPI—myocardial perfusion imaging; gBPS—gated blood-pool SPECT; CRT—cardiac resynchronization therapy; CRT-D—CRT devices with the defibrillation function.

#### *2.10. Statistical Analysis*

Statistical analysis was performed using the software package Statistica 10.0, StatSoft (USA). The Shapiro–Wilk test was used to assess the normality of the distribution of the trait. For variables with the normal distribution, the mean value (M) and standard deviation (SD) were calculated, while for others, the median (Me) with an interquartile range [*Q*1, *Q*3] were calculated. The nonparametric Mann–Whitney test for independent samples and the Wilcoxon test for dependent samples were used. The nonparametric Spearman analysis was used to assess the correlations between the pairs of quantitative features. Efficacy analysis was performed using logistic regression analysis. The forward-stepwise logistic regression analysis was used to evaluate the independent predictors of CRT response. Receiver operating characteristic (ROC) analysis was used to determine the diagnostic efficiency of the method using MedCalc statistical software package. We considered significant *p* values <0.05.

#### **3. Results**

#### *3.1. Patients' Baseline Clinical and Follow-Up Characteristics*

A total of 56 (100.0%) patients who underwent MPI, gBPS, CRT-D implantation, and 6-month follow-up were included in this study. The first group consisted of 45 (80.3%) individuals with a response to CRT (RESP), and the second group comprised 11 (19.7%) patients without response (non-RESP). The baseline demographics and clinical characteristics of the included patients are shown in Table 1. Considering all patients, the mean age was 57.0 ± 11.5 years, and 35 (62.5%) patients were males. The baseline 6MWDT, HF class NYHA, QRS duration, TTE parameters, arrhythmias before CRT-D implantation, LV

lead position, HF etiology, comorbid pathologies, and medical therapy records between the groups are also shown in Table 1. There were no significant differences between the groups in baseline demographics and clinical characteristics.

**Table 1.** Baseline demographics and clinical characteristics of enrolled patients.



**Table 1.** *Cont.*

Values are mean ± SD and Me [Q1; Q3] for continuous variables and *n* (%) for categorical variables. ACEI—angiotensin-converting enzyme inhibitors, GFR—glomerular filtration rate, non-RESP—patients without response to cardiac resynchronization therapy, RESP—patients with response to cardiac resynchronization therapy, VT—ventricular tachycardia, NYHA—New York Heart Association, LV—left ventricular, LVEF—left ventricular ejection fraction, LVESV—left ventricular end-systolic volume, AV—atrioventricular.

Both groups were comparable in terms of pre-CRT-D implantation scintigraphic parameters, except SRS, RV PFR, and MD indicators, assessed via MPI and gBPS. In RESP patients, the right ventricular peak filling rate (RV PFR) (*p* = 0.005), the left ventricular anterior-wall entropy (LV AW\_E) (*p* = 0.001), the left ventricular anterior-wall standard deviation (LV AW\_SD) (*p* = 0.0001), the right ventricular free-wall standard deviation (RV FW\_SD) (*p* = 0.011), the left ventricular entropy (LV\_E) (*p* = 0.033), and IVD (*p* = 0.022) were significantly higher, and SRS (*p* = 0.018) and RV PFR2 (*p* = 0.028) were significantly lower than in non-RESP subjects. A detailed comparison of pre-CRT-D implantation scintigraphic parameters is presented in Table 2.

Both groups were comparable in terms of TTE parameters before CRT-D implantation. In RESP patients 6 months after CRT-D implantation, LVESD (*p* = 0.015), LVEDD (*p* = 0.04), LVESV (*p* = 0.005), LVESI (*p* = 0.019), and LVSI (*p* = 0.02) were significantly lower, and LVEF (*p* < 0.001) was significantly higher than in non-RESP patients. The QRS duration 6 months after CRT-D implantation did not differ significantly between the groups. In RESP patients, QRS duration was 140.0 ms [130.0; 140.0], and non-RESP was 140.0 ms [130.0; 140.0] (*p* = 0.243).

Both groups were comparable in terms of the number of I (*p* = 0.215) and II (*p* = 0.332) HF class NYHA patients after 6 months of CRT-D implantation. The quantity of III HF class NYHA patients in the non-RESP (*n* = 7 (63.6%)) group was significantly higher than in the RESP (*n* = 9 (20.0%)) group (*p* = 0.026). Additionally, 6MWDT in the RESP (382.4 ± 88.1 m) group was significantly higher than in the non-RESP (288.1 ± 73.1 m) group (*p* = 0.003) after 6 months of CRT-D implantation.


**Table 2.** Scintigraphic characteristics before CRT-D implantation.

Values are expressed as Me [Q1; Q3]. IVD—interventricular dyssynchrony, LV HBW—left ventricular histogram bandwidth, LV PE—left ventricular phase entropy, RV HBW—right ventricular histogram bandwidth, RV PE—right ventricular phase entropy, RV FW\_SD—right ventricular free-wall standard deviation, LV S\_SD—left ventricular septal-wall standard deviation, LV S\_E—left ventricular septal-wall entropy, LV AW\_SD—left ventricular anterior-wall standard deviation, LV AW\_E—left ventricular anterior-wall entropy, LV LW\_SD—left ventricular lateral-wall standard deviation, LV LW\_E—left ventricular lateral-wall entropy, LV IW\_SD—left ventricular inferior-wall standard deviation, LV IW\_E—left ventricular inferior-wall entropy, LV PER—left ventricular peak emptying rate, LV PFR—left ventricular peak filling rate, LV PFR2—left ventricular peak filling rate on the second peak, RV PER—right ventricular peak emptying rate, RV PFR—right ventricular peak filling rate.

#### *3.2. Events According to the CRT-D Interrogation Data*

During 6 months of follow-up, VT events were registered in seven (12.5%) patients from both groups. In the first group (RESP patients), VT was registered in four (7.1%) patients: three unsustained VT with spontaneous termination and one sustained VT, terminated using ATP therapy. In the second group (non-RESP patients), VT was registered in three (5.4%) cases (*p* = 0.353). There were only three unsustained VT with spontaneous termination (*p* = 0.297). There were no lead dysfunction, dislocation, and inappropriate ICD therapy in either group. In RESP patients 6 months after device implantation, according to the CRT-D interrogation data, the paced atrioventricular delay was 148.7 ± 14.5 ms, the sensed atrioventricular delay was 110.0 ± 16.1 ms, the interventricular delay (first LV) was 25.0 ms [15.0; 40.0], and the time of biventricular pacing was 98.1 ± 1.9%. In non-RESP individuals 6 months after CRT-D implantation, the paced atrioventricular delay was

153.6 ± 18.0 ms (*p* = 0.464), the sensed atrioventricular delay was 116.8 ± 18.2 ms (*p* = 0.215), the interventricular delay (first LV) was 20.0 ms [10.0; 40.0] (*p* = 0.885), and the time of biventricular pacing was 98.7 ± 0.4% (*p* = 0.433).

#### *3.3. CRT Response Predictors*

According to the ROC analysis, the SRS and MD indicators (RV PFR, LV AW\_E, LV AW\_SD, RV FW\_SD, LV\_E, IVD, and RV PFR 2) were statistically significant predictors of CRT response (Table 3).


**Table 3.** The ROC analysis results.

Notes: 95% CI—95% confidence interval, AUC—area under the curve, IVD—interventricular dyssynchrony, LV AW\_E—left ventricular anterior-wall entropy, LV AW\_SD—left ventricular anterior-wall standard deviation, LV\_E—left ventricular entropy, RV FW\_SD—right ventricular free-wall standard deviation, RV PFR—right ventricular peak filling rate, RV PFR 2—right ventricular peak filling rate on the second peak, SRS—summed rest score.

The multivariate logistic regression analysis included the following indicators: QRS duration, gender, HF etiology, LV lead position, 6 min walk distance test, LVESV, LVEF, SRS, and MD. This analysis showed that only myocardial perfusion defects and MD indicators assessed via MPI and gBPS such as LV AW\_SD (OR 1.5275; 95% CI 1.1472–2.0340; *p* = 0.0037) and SRS (OR 0.7299; 95% CI 0.5627–0.9469; *p* = 0.0178) and also HF with a nonischemic etiology (OR 20.1425; 95% CI 1.2719–318.9961; *p* = 0.0331) were the independent predictors of CRT response.

The CRT response prediction model was performed considering the combination of LV AW\_SD, SRS, and the presence of nonischemic HF etiology. When ROC analysis was performed with the prediction model, the AUC was 0.949, Sen = 95.56, Spe = 81.82, and *p* < 0.001 (Figure 2).

The pairwise comparison of CRT response probability indicators revealed that the ROC curve with SRS (*p* = 0.001), HF nonischemic etiology (*p* < 0.001), the mechanical dyssynchrony indicators assessed via MPS and gBPS (IVD (*p* = 0.01), LVE (*p* < 0.001), RV PFR 2 (*p* = 0.003), RV PFR (*p* = 0.005), LV AW\_E (*p* = 0.037), and RV FW\_SD (*p* = 0.014)) ROC curves showed significant differences, with the exception of LV AW\_SD (*p* = 0.089) (Table 4).

**Table 4.** The ROC curves' pairwise comparison.


**Table 4.** *Cont.*


Notes: 95% CI—95% confidence interval, AD—area difference, CRT RPI—cardiac resynchronization therapy response probability indicator, IVD—interventricular dyssynchrony, LV AW\_E—left ventricular anterior-wall entropy, LV AW\_SD—left ventricular anterior-wall standard deviation, LV\_E—left ventricular entropy, RV FW\_SD—right ventricular free-wall standard deviation, RV PFR—right ventricular peak filling rate, RV PFR 2—right ventricular peak filling rate on the second peak, NIHF—nonischemic heart failure, SRS—summed rest score, MSE—mean squared error.

**Figure 2.** CRT response probability indicator, calculated according to the predictive model. AUC—area under the curve, Sen—sensitivity, Spe—specificity, t—threshold value.

#### **4. Discussion**

In the present study, it was shown that mechanical dyssynchrony (left ventricular anterior-wall standard deviation) and the scar burden assessed via cardiac scintigraphy in patients with HF of different etiologies may have prognostic value. Thus, HF with a nonischemic etiology, a low value of SRS, and a high value of LV AW\_SD are favorable predictive indicators of CRT response. This may be due to the smaller size of the scar and the involvement of the anterior LV wall in mechanical dyssynchrony.

The nonischemic origin of HF was previously reported to predict reverse remodeling in patients with more advanced HF symptoms. According to Ypenburg et al., CRT responders and patients with super responses more frequently had a nonischemic etiology of HF [35]. In a study by Verhaert et al., it was shown that the female gender and the nonischemic etiology of HF were associated with a much greater initial response to CRT [36]. Another study

by Said et al. found that women showed a greater echocardiographic response to CRT at 6 months of follow-up [37]. However, after the adjustment for body surface area and ischemic etiology, no differences were found in LV function measures or survival, suggesting that having a nonischemic etiology is responsible for greater response rates in women treated with CRT [37]. In our study, the number of patients with ischemic HF and nonischemic HF did not significantly differ in the groups of responders and nonresponders (*p* = 0.124). However, multivariate logistic regression showed that the nonischemic etiology of HF is an independent predictor of CRT response (95% CI 1.2719–318.9961; *p* = 0.0331). Furthermore, the subanalyses of multiple prospective randomized studies, including the Multicenter InSync Randomized Clinical Evaluation (MIRACLE), Cardiac Resynchronization—Heart Failure (CARE-HF), and Multicenter Automatic Defibrillator Implantation with Cardiac Resynchronization Therapy (MADIT-CRT), have confirmed the finding of greater reverse remodeling in HF with a nonischemic etiology [28,38].

The clinical significance of the scar burden assessed with MPI was demonstrated in several studies. Thus, in a study by Adelstein E. et al., it was found that among ischemic cardiomyopathy patients, lesser scar burden evaluated via SPECT MPI (SRS < 27) was associated with more favorable survival and reverse remodeling following CRT, with outcomes similar to nonischemic cardiomyopathy patients [27]. High scar burden (SRS ≥ 27) was associated with the lack of LV functional improvement, the absence of reverse remodeling, and worse survival [27]. In our study, patients with HF of different etiologies were included, and a comparison of the groups with ischemic and nonischemic etiologies was not performed. In patients with HF, myocardial perfusion impairment can vary widely and may be associated with the presence of myocardial fibrosis due to myocardial infarction in ischemic HF patients, or the presence of interstitial fibrosis, which may play a crucial role in the process of myocardial remodeling [39]. Recent studies revealed that myocardial fibrosis is an independent predictor of mortality and morbidity in patients with dilated cardiomyopathy undergoing CRT [40]. Additionally, in our study, which included mixed samples of patients with ischemic and nonischemic etiologies, the SRS assessed using MPI ≤ 7.0 was an independent significant predictor for CRT patient selection (95% CI 0.5627–0.9469; *p* = 0.0178).

The predictive value of LV mechanical dyssynchrony for CRT patient selection measured via MPI and gBPS was widely studied. In a study with 142 CRT patients, LV mechanical dyssynchrony parameters such as the systolic histogram bandwidth (95% CI 0.98–1.00, *p* = 0.041), the diastolic-phase standard deviation (95% CI 0.94–1.00, *p* = 0.041), and the diastolic histogram bandwidth (95% CI 0.98–1.00, *p* = 0.028) were significant independent predictors of CRT response only for nonischemic HF patients [25]. For ischemic HF individuals, none of the LV mechanical dyssynchrony parameters were statistically significant [25]. In a study by Henneman M. et al. with 42 CRT patients, the ROC analysis showed that the optimal cut-off values for the phase standard deviation and histogram bandwidth were 43◦ (sensitivity and specificity of 74%) and 135◦ (sensitivity and specificity of 70%), respectively [41]. In a study with 324 consecutive individuals with nonischemic HF CRT patients, it was shown that the systolic-phase standard deviation, adjusted to age, hypertension, diabetes, aspirin, betablockers, diuretics, QRS, and LVEF, was an independent predictor of all-cause mortality (HR 1.97, 95% CI 1.06–3.66, *p* = 0.033) [42]. In our study, the responders and nonresponders did not differ in terms of such indicators of global LV mechanical dyssynchrony as the phase standard deviation and histogram bandwidth, and these indicators had no prognostic value, but the LV phase entropy was higher in the responders' sample. However, the novelty of our study was the use of gBPS for assessing the regional mechanical dyssynchrony indicators separately for the septum, anterior, posterior, and lateral walls of LV. These indicators (LV AW\_E, LV AW\_SD, and RV FW\_SD) with IVD and contractile function indicators (RV PFR and RV PFR 2) were statistically significant predictors of CRT response. The multivariate logistic regression showed that LV AW\_SD (OR 1.5275; 95% CI 1.1472–2.0340; *p* = 0.0037) was the independent predictor of CRT response. This may indicate that the assessment of regional myocardial dyssynchrony may provide additional information for successful resynchronization therapy. However, disagreements with previous studies underline the importance of these findings and the need for future large-scale studies.

There are few publications about the combined assessment of the factors associated with reverse remodeling, which can be used for the improved selection of patients for CRT. The significance of the CRT response score was shown in a major randomized trial MADIT-CRT (*n* = 1761) [28]. This score included seven factors associated with the echocardiographic response to CRT-D and made up the response score (female sex, nonischemic HF, LBBB, QRS duration ≥150 ms, prior hospitalization for HF, left ventricular end-diastolic volume ≥125 mL/m<sup>2</sup> and left atrial volume <40 mL/m2). The multivariate analysis showed a 13% (*p* < 0.001) increase in the clinical benefit of CRT-D per one-point increment in the response score (range, 0–14) and a significant direct correlation between the risk reduction associated with CRT-D and the response score quartiles: Patients in the first quartile did not reveal a significant reduction in the risk of HF or death with CRT-D (hazard ratio = 0.87; *p* = 0.52); patients in the second and third quartiles had 33% (*p* = 0.04) and 36% (*p* = 0.03) risk reductions, respectively; and patients in the upper quartile experienced a 69% (*p* < 0.001) risk reduction (*p* = 0.005). In our study, the multivariate logistic regression with inclusion factors such as QRS duration, gender, HF etiology, LV lead position, 6 min walk distance test, LVESV, LVEF, SRS, and mechanical dyssynchrony indicators assessed via gBPS showed that LV AW\_SD, SRS, and HF with a nonischemic etiology were the independent predictors of CRT response (95% CI 0.856–0.990; AUC = 0.949; Sen = 95.56; Spe = 81.82; *p* < 0.001).

According to a study by Forleo et al., quadripolar leads allowed for nonapical pacing with acceptable electrical parameters in the majority of CRT recipients, although differences were found among the currently available devices [43]. In our study, quadripolar and bipolar LV leads from various manufacturers were implanted. A detailed analysis of lead characteristics between different companies was not performed. However, a comparison between the groups of responders and nonresponders to CRT in terms of the number of implanted quadripolar and bipolar LV leads was made. The comparative analysis between the groups did not reveal significant differences (*p* = 0.392).

Thus, the predictive model with the combination of LV AW\_SD, SRS, and the presence of nonischemic HF etiology has a high prognostic value and can be used as an additional CRT response predictor in patients with HF of different etiologies.

#### *Study Limitations*

The limitations of this study included its relatively small sample size and short followup period; the left ventricular mechanical dyssynchrony indicators were not determined in terms of their dynamics, and this was a nonrandomized, single-center study.

#### **5. Conclusions**

In patients with HF of ischemic and nonischemic etiologies, the left ventricular mechanical dyssynchrony assessed via gBPS may be useful in identifying responders to CRT. In these patients, a combined assessment of the factors associated with reverse remodeling (HF with a nonischemic etiology, LV AW\_SD, and SRS) can be used for the improved selection of patients for CRT.

**Author Contributions:** Conceptualization, T.A.A. and R.E.B.; Validation, M.S.K.; Formal analysis, M.S.K.; Investigation, T.A.A. and S.N.K.; Resources, T.A.A.; Data curation, T.A.A.; Writing – original draft, T.A.A. and M.S.K.; Writing – review & editing, M.S.K., A.I.M., R.E.B., S.I.S., S.N.K., V.V.S., Y.V.V., K.V.Z. and S.V.P.; Visualization, S.I.S. and Y.V.V.; Supervision, R.E.B., S.N.K. and S.V.P.; Project administration, R.E.B., K.V.Z. and S.V.P. All authors were responsible for the inclusion of patients and were involved in the critical manuscript revision for important intellectual content. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

**Institutional Review Board Statement:** This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Ethics Committee of Cardiology Research Institute, Tomsk National Research Medical Center, Russian Academy of Science (protocol code 163 and 8 November 2017).

**Informed Consent Statement:** Informed consent was obtained from all the subjects involved in this study. Written informed consent has been obtained from the patients to publish the paper in an anonymous form.

**Data Availability Statement:** According to the internal regulations of the Institute, all data are the property of the Institute and can only be provided anonymously after an official request.

**Conflicts of Interest:** No financial or other relations that could lead to a conflict of interest are associated with this study.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

### *Systematic Review* **Systematic Review on S-ICD Lead Extraction**

**Riccardo Vio 1, Enrico Forlin 1,2, Viktor Culi´ ˇ c3,4, Sakis Themistoclakis 1, Riccardo Proietti 5,\* and Paolo China <sup>1</sup>**


**Abstract:** Background and purpose: Subcutaneous implantable cardioverter defibrillators (S-ICDs) have emerged in recent years as a valid alternative to traditional transvenous ICDs (TV-ICDs). Therefore, the number of S-ICD implantations is rising, leading to a consequent increase in S-ICDrelated complications sometimes requiring complete device removal. Thus, the aim of this systematic review is to gather all the available literature on S-ICD lead extraction (SLE), with particular reference to the type of indication, techniques, complications and success rate. Methods: Studies were identified by searching electronic databases (Medline via PubMed, Scopus and Web of Science) from inception to 21 November 2022. The search strategy adopted was developed using the following key words: subcutaneous, S-ICD, defibrillator, ICD, extraction, explantation. Studies were included if they met both of the following criteria: (1) inclusion of patients with S-ICD; (2) inclusion of patients who underwent SLE. Results: Our literature search identified 238 references. Based on the abstract evaluation, 38 of these citations were considered potentially eligible for inclusion, and their full texts were analyzed. We excluded 8 of these studies because no SLE was performed. Eventually, 30 studies were included, with 207 patients who underwent SLE. Overall, the majority of SLEs were performed for non-infective causes (59.90%). Infection of the device (affecting either the lead or the pocket) was the cause of SLE in 38.65% of cases. Indication data were not available in 3/207 cases. The mean dwelling time was 14 months. SLEs were performed using manual traction or with the aid of a tool designed for transvenous lead extraction (TLE), including either a rotational or nonpowered mechanical dilator sheath. Conclusions: SLE is performed mainly for non-infective causes. Techniques vary greatly across different studies. Dedicated tools for SLE might be developed in the future and standard approaches should be defined. In the meantime, authors are encouraged to share their experience and data to further refine the existing variegated approaches.

**Keywords:** subcutaneous implantable cardioverter defibrillator; S-ICD; ICD; extraction; explantation; infection

#### **1. Introduction**

Since their development, Implantable Cardioverter Defibrillators (ICDs) have been a fundamental part of the primary and secondary prevention of sudden cardiac death (SCD) [1]. Although transvenous ICDs (TV-ICDs) are still the standard of care, their longterm use is associated with a variety of device-related complications, such as infections or lead displacement and rupture, in some cases requiring surgical revision or even the extraction of the entire device [2,3]. Subcutaneous ICDs (S-ICDs) thus became a valid alternative to reduce the occurrence of adverse events related to the presence of intravenous leads, especially in young patients with a long life expectancy, normal heart, and no need for

**Citation:** Vio, R.; Forlin, E.; Culi´ ˇ c, V.; Themistoclakis, S.; Proietti, R.; China, P. Systematic Review on S-ICD Lead Extraction. *J. Clin. Med.* **2023**, *12*, 3730. https://doi.org/10.3390/ jcm12113710

Academic Editor: Simon Rabkin

Received: 24 April 2023 Revised: 20 May 2023 Accepted: 25 May 2023 Published: 27 May 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

pacing or cardiac resynchronization therapy, at increased risk of infections and with limited vascular access [4,5]. Since S-ICDs have consistently been proved to be effective in terminating malignant arrhythmias, the implantation of such devices is rising. In parallel with the growing number of S-ICDs implanted, there is an absolute increase in S-ICD-related complications (e.g., infection) that may require complete system removal [6–8]. Although transvenous lead explantation (TLE) is a relatively common procedure, with several approaches, dedicated tools and specific guidelines [9], S-ICD lead extraction (SLE) still relies on poor clinical data regarding its technical execution (with simple manual traction as the only established method), efficacy and safety [10,11]. This could be particularly relevant not only for the increasing necessity of extraction procedures, but also for the growing dwelling time of the S-ICDs implanted, typically associated with the development of fibrotic adherences and calcifications around the lead that could interfere with extraction [10]. Thus, the aim of this systematic review is to gather all the available literature on SLE, with particular reference to type of indication and the method used to perform the extraction.

#### **2. Materials and Methods**

#### *2.1. Design*

This systematic review was conducted and reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations [12]. The systematic review was not registered.

#### *2.2. Study Selection*

Studies were identified by searching electronic databases (Medline via PubMed, Scopus and Web of Science) from inception to 21 November 2022. The literature search used text and relevant indexing to capture data on S-ICD lead extraction. The search strategy adopted was developed using the following key words: subcutaneous, S-ICD, defibrillator, ICD, extraction, explantation. The string adopted for PubMed was ("subcutaneous" OR "S-ICD") AND ("defibrillator" OR "ICD") AND ("extraction" OR "explantation"). Only studies in English were included in the review. Studies were included if they met both of the following criteria: (1) inclusion of patients with S-ICD; (2) inclusion of patients who underwent SLE.

#### *2.3. Data Extraction*

Information was extracted from each included study regarding the (1) baseline characteristics of patients and country where the study was performed; (2) indications for SLE; (3) dwelling time; (4) technique used for SLE; (5) complications; (6) success; and (7) management after SLE. Two authors (R.V. and E.F.) independently extracted data from studies and entered them into the data extraction form. Disagreements were resolved by discussion; if no accord was reached, it was planned that a third author (P.C.) would decide.

#### *2.4. Quality Assessment*

The JBI tool was used to perform the quality assessment of the included studies [13]. Supplementary Materials S1 reports how studies were rated.

#### **3. Results**

#### *3.1. Study Selection*

Our literature search identified 238 references (Figure 1). Based on the abstract evaluation, 38 of these citations were considered potentially eligible for inclusion, and their full texts were analyzed in more detail [10,11,14–49]. We excluded 8 of these studies because no SLE was performed [10,18,19,21,27,29,46,47]. Eventually, 30 studies were included [10,11,14–17,20,22–26,28,30–41,43–45,48,49].

**Figure 1.** Flow chart of the literature search.

#### *3.2. Characteristics of the Studies*

Overall, 207 patients were included in this analysis. One patient underwent SLE twice [43]. Of the 30 included studies, the majority (16/30) were retrospective cohort studies [15–17,22,23,26,30,33,36–38,41,43,44,48,49]. Five were prospective studies [14,31,32,34,35], and the remaining nine were case reports [10,11,20,24,25,28,39,40,45]. The number of patients included in the studies ranged from 1 to 1637. The full study characteristics are summarized in Table 1.

**Table 1.** Characteristics of the included studies.



#### **Table 1.** *Cont.*

S-ICD, subcutaneous implantable cardioverter defibrillator; TLE, transvenous lead extraction; TV-ICD, transvenous implantable cardioverter defibrillator; PGR, pulse generator replacement; Brs, Brugada Syndrome; DCM, dilated cardiomyopathy; PM, pacemaker; CIED, cardiovascular implantable electronic device; ICM, ischemic cardiomyopathy; AC, arrhythmogenic cardiomyopathy; SQC, subcutaneous shocking coils; pt, patient; pts, patients. \* 1 pt underwent extraction twice.

#### *3.3. Indications for S-ICD Extraction*

Overall, the majority of SLEs were performed for non-infective causes (59.90%). Among these, the occurrence of inappropriate shocks was responsible for 16.91% of total SLEs, necessity for cardiac resynchronization therapy (CRT) was responsible for 8.70%, progression of disease to heart transplantation or left ventricular assist device (LVAD) implantation was responsible for 7.25% and sensing issues were responsible for 4.35%. Infection of the device (affecting either the lead or the pocket) was the cause of SLE in 38.65% of cases. Indication data were not available in 3/207 cases [17,22]. The full list of all the indications is reported in Table 2.

#### *3.4. Dwelling Time*

The mean time from implantation to extraction (i.e., dwelling time) of S-ICDs was 14 months, ranging from 30 days [37] to over 8 years [36]. The dwelling time was not available in 10 studies [14–16,20,22,23,32,35,38,44]. The majority of studies in which this information was reported showed a mean time of one year or less [24,26,33,36,39,43,45,48]. The dwelling time was particularly short in studies where infections were the indication for more than 50% of SLEs performed [31,36,37,43].


**Table 2.** Indications for SLE.

CRT, cardiac resynchronization therapy; LVAD, left ventricular assist device; MRI, magnetic resonance imaging.

#### *3.5. Method Used for SLE*

In the studies analyzed, SLE required a two- or three-incision technique, depending on the way the lead was originally inserted and on eventual complications. The first incision was performed generally along the midaxillary line, in order to open the pocket and remove the pulse generator, freeing also the proximal part of the lead. The second incision was then executed on the region of xiphoid apophysis, fundamental to release the suture anchoring the lead and usually exploited as the access to extract the proximal and distal ends of the lead. An additional incision could be performed at the level of the manubriosternal junction in order to reach the distal coil of the lead, cutting eventual sutures that anchored it to the periosteal fascia or releasing it from fibrous tissue. Manual traction was then applied to unthread the lead, generally from subxiphoid access. In some cases, typically characterized by a long dwelling time, fibrosis of the tissues around the lead offered resistance to its removal, requiring the use of a mechanical sheath to relieve it [9,10,18,24,26]. This tool, originally designed for TLE, is manufactured to be slipped around the lead and then, with traction and counter-traction and simultaneous rotation, progressively advanced over the catheter. This technique allows the operator to dilatate and to break off fibrotic tissue along the lead, favoring its removal. Additional treatment could then be executed according to the S-ICD extraction's indication (e.g., local antimicrobial treatment in case of pocket infection). The extraction technique was described in 11 studies [10,11,17,20,24–26,28,36,40,41], a minority of those included in this analysis. Among these studies, 7/11 were case reports [10,11,20,24,25,28,40]. Table 3 reports the methods used for SLE in our study population.

**Table 3.** Methods used for SLE.


SLE, S-ICD lead extraction.

#### *3.6. Complications and Success Rate*

There were no periprocedural complications reported during SLE. This point was specifically addressed by the two included studies reporting the greatest numbers of SLEs [17,26]. Pothineni et al. reported that in their cohort of 64 patients who underwent SLE, no complications occurred [17]. The secondary endpoint of the study by Behar et al. [26] included procedural complications, and again, no procedure-related complications were reported.

Only 1 case of procedural failure occurred among the 207 reported because the lead could definitely not be extracted [26].

#### *3.7. Management after SLE*

Table 4 reports which management options were chosen after SLE, including the percentages of S-ICD or TV-ICD reimplantation.


**Table 4.** Management after SLE.

ICD, subcutaneous implantable cardioverter defibrillator; TV-ICD, transvenous implantable cardioverter defibrillator. \* 1 pt was reimplanted twice.

Of note, in the study by Brouwer et al., one patient underwent a second S-ICD extraction after SLE during subsequent follow-up for recurrent infection [43].

#### **4. Discussion**

To the best of our knowledge, our study is the first systematic review on SLE.

According to our systematic review, the majority of SLEs are performed for noninfective causes (59.90%). These data differ from those for TLEs, which are performed mainly for infective causes (52.8%) [50]. Among the non-infective indications for SLE, inappropriate shocks accounted for the most part (16.91%). Moreover, sensing issues accounted for 4.35% of cases. Despite rare cases of undersensing of ventricular fibrillation [17], sensing issues usually include refractory oversensing due to myopotential, P- or T-wave oversensing or R-wave double counting [30]. Eventually, oversensing may lead to inappropriate shocks. Inappropriate shocks are a well-known drawback of S-ICDs, significantly affecting patients' quality of life [51]. However, recent data showed a consistent reduction in inappropriate shocks by S-ICDs (3.1% in 1 year) with the use of high-rate cutoffs, as well as current generation electrogram filtering and discrimination algorithms [52]. Additionally, exercise tests before S-ICD implantation, especially in patients with Brugada syndrome, may further reduce the incidence of future inappropriate shocks [53]. Given all the abovementioned improvements, we can expect a decrease for such indications in future studies on SLE.

A necessity for resynchronization therapy or pacing is among other prominent causes for SLE. Even though it usually cannot be predicted at implantation, accurate selection of candidates for S-ICD implantation must be performed to minimize this possible scenario.

The dwelling time was very different in our systematic review on SLE compared to that reported for TLE. In fact, our study showed a mean time from implant to extraction of 14 months, compared to a mean dwelling time of 6.4 years in the ELECTRA study on TLE [50]. This discrepancy can be explained by several factors. First, in the studies included in our systematic review, many authors included patients in which the device was removed within a year and with simple manual traction; these procedures are better referred to as "explantation", according to the terminology used for transvenous PM/ICDs. Other authors did not mention at all the dwelling time or the approach used for extraction, but we decided to include all studies in order to gather all the available literature in the field. Secondly, S-ICDs have been available since the early 2010s, whereas transvenous PM/ICDs were released on the market several years prior, and long-term follow-up data are available. Lastly, infections on transvenous PM/ICDs may occur as part of a systemic infection/bacteremia with the source of infection coming from a different site. In the case of S-ICDs, infections are primarily local and related to the implantation procedure; therefore, they tend to occur earlier. In fact, we found that the dwelling time was particularly short in studies where infections were the indication for more than 50% of SLEs [31,36,37,43].

Regarding techniques for SLE, simple manual traction is the easier way, and it was sufficient to remove S-ICD leads in many of the cases reported (99/207) (Table 3). However, this approach cannot always be pursued due to fibrosis surrounding the parasternal coil. According to our systematic review, in 14/207 cases, tools were used to complete a successful SLE. Patel et al. were the first authors to report the successful use of a rotating mechanical dilator sheath (TightRail, Spectranetics) for SLE [28]. Later on, Allison et al. replicated an SLE using the rotating mechanical dilator sheath (TightRail, Spectranetics), but also with the aid of a bulldog lead extender (Cook Medical) [20]. Migliore et al. reported the use of a non-powered mechanical dilator sheath (LR-TSS-11.0, Cook Medical) to disrupt fibrotic adhesions around the coil and the distal tip of the S-ICD lead [10]. The lead was then retracted into the sheath and successfully extracted in the absence of any complications. These studies highlight the absence of a common strategy and of dedicated tools for SLE, which might be developed in the future.

After SLE, a large number of patients (75/207) underwent TV-ICD reimplantation (Table 4). A minor part of the cohort underwent S-ICD reimplantation (25/207). Of note, one patient was extracted and reimplanted twice with an S-ICD because of recurrent local infection [43]. This event highlights the importance of a waiting period before reimplantation to allow complete healing from infection. In this scenario, wearable ICDs may act as a bridge therapy. In the study by Van der Stuijt et al. [23], the authors provided a patient with a wearable ICD after SLE for 3 months, allowing the resolution of infection with antibiotics before S-ICD reimplantation.

#### *Limitation*

Our study has several limitations. The results are based primarily on retrospective studies and case reports, and we acknowledge the risk of overlapping cohorts among different studies. Furthermore, as discussed above, no clear distinction was made between explantation and extraction, and all cases of S-ICD lead removal were collectively considered as SLE. Finally, we could not exclude possible bias due to overlapping cohorts between some of the included studies.

#### **5. Conclusions**

According to our systematic review, the causes for SLE are primarily non-infective, and the dwelling time seems to be lower compared to that in the literature data on TLE. Techniques vary greatly across different studies, including manual traction and the use of either rotational or non-powered mechanical sheaths. Dedicated tools for SLE might be developed in the future, and standard approaches should be defined. In the meantime, authors are encouraged to share their experience and data to further refine the existing variegated approaches.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jcm12113710/s1. Supplementary Material: Quality Assessment.

**Author Contributions:** Conceptualization, R.V., R.P. and P.C.; methodology, R.V. and R.P.; data curation, R.V., E.F. and P.C.; writing—original draft preparation, R.V. and E.F.; writing—review and editing, R.V., E.F., V.C., S.T., R.P. and P.C. All authors have read and agreed to the published version ˇ of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** No new data were created or analyzed in this study. Data sharing is not applicable to this article.

**Acknowledgments:** The authors wish to thank Erika Borsetto for assisting in the preparation of the figure.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
