Unexpected Procedure Difficulties Increasing the Complexity of Transvenous Lead Extraction: The Single Centre Experience with 3721 Procedures

Background: Transvenous lead extraction (TLE) should be completed, even when facing difficulties which have yet to be described. The aim was to explore unexpected TLE obstacles (the circumstances of the occurrence and influence on TLE outcome). Methods: The retrospective analysis of a single centre database containing 3721 TLEs. Results: Unexpected procedure difficulties (UPDs) occurred in 18.43% of cases (singles in 12.20% of cases and multiples in 6.26% of cases). These included blockages in the lead venous approach in3.28% of cases, functional lead dislodgement in 0.91% of cases, and loss of broken lead fragment in 0.60% of cases. All of them, including implant vein—in 7.98% of cases, lead fracture during extraction—in 3.84% of cases, and lead-to-lead adherence—in 6.59% of cases, Byrd dilator collapse—in 3.41% of cases, including the use of an alternative prolonged the procedure but had no influence on long-term mortality. Most of the occurrences were associated with lead dwell time, younger patient age, lead burden, and poorer procedure effectiveness and complications (common cause). However, some of the problems seemed to be related to cardiac implantable electronic devices (CIED) implantation and the subsequent lead management strategy. A more complete list of all tips and tricks is still required. Conclusions: (1) The complexity of the lead extraction procedure combines both prolonged procedure duration and the occurrence of lesser-known UPDs. (2) UPDs are present in nearly one fifth of the TLE procedures, and can occur simultaneously. (3) UPDs, which usually force the extractor to expand the range of techniques and tools, should become part of the training in transvenous lead extraction.


Introduction
Transvenous lead extraction (TLE) plays a key role in the management of patients with cardiac implantable electronic devices (CIED). It is a highly effective procedure (over 95%) with an acceptable rate of major complications (1.6-2.5%) [1][2][3][4]. Its goal is to remove the lead in its entirety, with a minimal risk of major complications (depending mainly on the operator's experience and the utilised tools), and without complication-related deaths (regarding the organization of the procedure) [1][2][3][4]. The procedure has to be successful and completed from start to finish despite the occurrence of unforeseen difficulties (e.g., the fracture of the targeted lead).
The prediction and management of major complications has been the subject of several reports [5][6][7][8]. Increased procedure difficulty was defined as prolonged extraction or fluoroscopy time [9][10][11][12][13][14][15][16], the use of advanced tools and methods [12][13][14][15][16], or as the increased which the broken sheath is blocked and cannot be moved relative to the lead (to varying degrees); and (2) A delayed BDF when twisting the tube blocks the lead. The sheath can be easily broken into two pieces if pulled or moved forward. BDF (not a complication unless a serious clinical event that may be life-threatening ensues) always requires special and careful sheath manoeuvres (tube removal or shift into the straight segment of the vein), also pulling the lead blocked by the broken sheath fragment, without countertraction, which may cause major procedural complications [21,22] (Figure 1).

Figure 1.
Byrd dilator collapse/fracture. Polypropylene catheter kinking is not a problem if detected early (A-C). Each successive rotation of the kinked polypropylene catheter rolls it up and tightens on the lead (E-H), making it more and more difficult to remove (D). Good quality fluoroscopy and careful observation of the entire working catheter can prevent secondary major problems [22].
Lead-to-lead adherence is when mechanical sheaths fail to disrupt adhesions between two leads, as when one lead is freed, the other lead is pulled down, or is wrapped around the lead being extracted [21,[23][24][25][26].
Lead fracture is defined as the complete break into two parts to be removed separately. The proximal part was easily removed using the dilator sheath. The distal lead part (>4 cm) or fragment (<4 cm) was removed in the next stage of the procedure [21,[26][27][28][29][30][31][32][33][34] (Figure 2). . Each successive rotation of the kinked polypropylene catheter rolls it up and tightens on the lead (E-H), making it more and more difficult to remove (D). Good quality fluoroscopy and careful observation of the entire working catheter can prevent secondary major problems [22].
Lead-to-lead adherence is when mechanical sheaths fail to disrupt adhesions between two leads, as when one lead is freed, the other lead is pulled down, or is wrapped around the lead being extracted [21,[23][24][25][26].
The loss of the broken lead fragment occurs when the main part of the lead is cut free and removed but both ends remain in place and a lead fragment is embolized into the pulmonary vascular bed. This difficulty group also included the loss of the silicone tube in the cardiovascular system. Both were removed in the next stage of the procedure [35][36][37][38][39].
Dislodgement of an active (functional) lead refers to the accidental displacement of the tip of the functional lead or a significant shift of the entire lead that made the venous route inaccessible. Such leads should be replaced, thus increasing the procedure time. We made every effort to avoid the displacement of leads that were difficult to reimplant (left ventricular and His bundle pacing) [40]. The use of an alternative venous approach refers to all procedures accomplished a vascular access site other than the implant vein [16,21,[24][25][26][28][29][30][31][32]34].
The loss of the broken lead fragment occurs when the main part of the lead is cut and removed but both ends remain in place and a lead fragment is embolized into pulmonary vascular bed. This difficulty group also included the loss of the silicone t in the cardiovascular system. Both were removed in the next stage of the procedure 39] Dislodgement of an active (functional) lead refers to the accidental displacemen the tip of the functional lead or a significant shift of the entire lead that made the ven route inaccessible. Such leads should be replaced, thus increasing the procedure time made every effort to avoid the displacement of leads that were difficult to reimplant ventricular and His bundle pacing) [40].
In the last 17 years, the organisation of lead extraction has evolved from proced performed in the electrophysiology laboratory using intravenous analgesia/seda [8,21] to procedures in the hybrid room in patients only under general anaesthesia. In last 7 years, the core extraction team has consisted of the same highly experien extractor (now usually serving as a proctor), experienced echocardiographer, dedicated cardiac surgeon [21,[25][26][27]. A total of 1188 (31.9%) TLE procedures w performed using transoesophageal echocardiographic monitoring, and the car surgeon was the co-operator during 1994 (53.59%) of the procedures. Some 1985 (53.3 procedures were performed in a hybrid room or in a cardiac surgery operating theatr
In the last 17 years, the organisation of lead extraction has evolved from procedures performed in the electrophysiology laboratory using intravenous analgesia/sedation [8,21] to procedures in the hybrid room in patients only under general anaesthesia. In the last 7 years, the core extraction team has consisted of the same highly experienced extractor (now usually serving as a proctor), experienced echocardiographer, and dedicated cardiac surgeon [21,[25][26][27]. A total of 1188 (31.9%) TLE procedures were performed using transoesophageal echocardiographic monitoring, and the cardiac surgeon was the co-operator during 1994 (53.59%) of the procedures. Some 1985 (53.35%) procedures were performed in a hybrid room or in a cardiac surgery operating theatre.

Dataset and Statistical Methods
We split the group into subgroups for the analysis of events and data patients. First, we grouped patients according to the occurrence of obstacles during lead extraction (unexpected procedure difficulties): blockage in the lead implant vein/subclavian region, polypropylene dilator collapse/fracture, lead-on-lead binding, lead fracture, the use of an alternative venous approach, the loss of broken lead fragment or functional lead dislodgement. As an event of interest, unexpected procedure difficulty could occur more than once in the same patient, and the total number of events in our analysis did not correspond to the number of extraction procedures. We analysed potential patient-related risk factors, CIED-related factors, technical problems during lead extraction, TLE complications, efficacy, survival after the procedure, and prognostic factors for the occurrence of UPDs. TLEs involving removal of broken leads with proximal ends in the cardiovascular system were excluded from analysis because the problem had been known before proceeding with the extraction, and therefore such situations did not meet the criteria for an unexpected problem [41].
The statistical significance of the impact of the assessed factors on the occurrence of individual UPDs was determined by comparing the group with a given UPD with the group of patients in whom an event of interest did not occur. All continuous variables are presented as means ± standard deviation. All categorical variables are presented as counts and percentages. Due to unequal sample sizes, the significance of differences was determined using the nonparametric Chi 2 test with Yates correction or the unpaired Mann-Whitney U test, as appropriate. A linear regression analysis was used to identify predictors of one UDP, more than one UDP, or any UDP occurrence. Variables with p-values < 0.05 under univariable analysis were entered into the multivariable models and presented in tables. To determine the impact of UDP on survival, the Kaplan-Meier survival curves were plotted, and were evaluated with the log rank test. A p-value less than 0.05 was considered significant. A statistical analysis was performed with Statistica 13.3 (TIBCO Software Inc.).

Approval of the Bioethics Committee
All patients gave their informed written consent to undergo TLE and use anonymous data from their medical records, as approved by the Bioethics Committee at the Regional Chamber of Physicians in Lublin, no. 288/2018/KB/VII. The study was carried out in accordance with the ethical standards of the 1964 Declaration of Helsinki.

Results
The study population was dominated by non-infectious indications (68.61%), including lead replacement (46.82%), caused by mechanical lead damage (26.93%), lead dysfunction (exit/entry block, extracardiac pacing) (22.55%) and change of pacing mode/upgrading, downgrading (6.26%) ( Table 1). We tried to identify patient-related risk factors for the occurrence of UPD. Table 2 shows that 81.53% of the procedures were trouble-free, one UPD occurred in 12.20% of the procedures, and multiple UPDs occurred in 6.26% of them. The most common UPDs were blockages in the lead implant vein/subclavian region (7.98%) and lead-on-lead binding (6.59%). Other unexpected problems were rare: Byrd dilator collapse/fracture (3.41%), fracture of the targeted lead (3.84%), use of an alternative venous approach (3.28%), functional lead dislodgement (0.91%), and the loss of a broken lead fragment (0.60%). Younger patient age at first system implantation similar to patient age during TLE increased the risk of Byrd dilatator collapse, fracture of the targeted lead, the use of an alternative venous approach and the occurrence of multiple or any UPD. There was no relationship between gender and the occurrence of UPDs. Patients with ischemic heart disease as underlying disease were at a lower risk of most UPDs, especially Byrd dilatator collapse, the use of an alternative venous approach, the loss of a broken lead fragment, and procedures with the occurrence of any technical problem. Higher LVEF values were associated with an increased risk of targeted lead fracture, the use of an alternative venous approach, and multiple UPDs. A low Charlson comorbidity index increased the risk of Byrd dilator collapse/fracture, targeted lead fracture, and the use of an alternative venous approach. * As unexpected procedure difficulty could occur more than once in the same patient, the total number of UPDs does not correspond to the number of extraction procedures. TLE-transvenous lead extraction, UPD-unexpected procedure difficulty, LVEF-left ventricular ejection fraction.
The study shows that the history of CIED and pacing are potential risk factors for increased procedure complexity and major complications. The number of leads in the heart before TLE similar to the presence of 4 and >4 leads in the heart before TLE increased the risk of lead-on-lead binding, the use of an alternative venous approach, the loss of broken lead fragments, and functional lead dislodgement. The presence of abandoned leads before TLE was associated with the more frequent occurrence of UPDs. Leads on both sides of the chest increased the risk of Byrd dilator collapse/fracture, the use of an alternative venous approach, and the loss of a broken lead fragment. The number of procedures before lead extraction in comparison to the other factors had a weak influence on the occurrence of UPDs. The presence of abnormally long lead loops in the atrium or in the ventricle before TLE was often associated with lead-on-lead binding, the fracture of the targeted lead, the use of an alternative venous approach, and the loss of a broken lead fragment. Importantly, in this study all potential risk factors were associated with the more frequent occurrence of multiple UPDs in the same patient.
In the next stage, we checked whether conventional risk factors for major complications of the procedure were also indicators of an increased risk of UPD. And as shown in Table 3, the removal of CIEDs with an ICD lead lowered the probability of UPDs, but the number of extracted leads per patient, similar to extraction of three or more leads, abandoned leads or passive leads, significantly increased the risk of all UPDs; the increase in risk was mainly due to Byrd dilator collapse/fracture, lead-on-lead binding, the fracture of targeted leads, the use of an alternative venous approach, and the loss of a broken lead fragment. It is obvious that the most important risk factor for major complications, namely lead dwell time, was strongly associated with the occurrence of all UPDs (except functional lead dislodgement). To sum up, there are at least four procedure-dependent risk factors for UPDs (the extraction of three or more leads, abandoned lead(s), and passive fixation lead(s), as well as long or very long implant duration).
In the next stage, we examined the relationship between the indicators of TLE complexity (extraction time, use of second-line tools and advanced tools) and the risk of UPD. The most representative indicator of procedure duration (defined by fluoroscopy time) was the extraction time of all leads and the average time of single lead extraction, since global procedure duration includes reimplantation of a new system in non-infectious cases. Thus, the lead extraction time was prolonged for all UPDs (mainly due to the fracture of the targeted lead, the use of an alternative venous approach, and the loss of a broken lead fragment). The occurrence of UPDs was associated with a more frequent use of secondline tools (mechanical rotational threaded tip) or advanced tools (lassos or Dormia basket catheters or other tools for a femoral approach).
The data presented in Table 4 shows that to some extent the predictors of MC also indicate an increased risk of UPDs (SAFeTY, EROS 2, 3). A similar direction of change (but less pronounced) was observed for MB score (the need for advanced tools to achieve TLE success), LED score (difficult TLE defined by fluoroscopy time), and advanced TLE (Mazzone) score (the need for advanced TLE techniques).
It seems that our recently developed [42] lead extraction difficulty score-LECOM score (combined lead extraction time, use of second-line or advanced tools and advanced techniques) is most useful for the prediction of UPDs.
Next, we analysed the relationship between the occurrence of major complications, incomplete lead extraction and procedural success, and the presence of UPDs. Patients with major complications (any), haemopericardium, emergent surgical intervention, and only partial radiographic success (retained lead tip or <4 cm lead fragment) were significantly more likely to have UPDs. These clear differences were not found for tricuspid valve damage during TLE, haemothorax, other major complications because there were too few cases to draw conclusions. Furthermore, similar to previous analyses, it should be underlined that all major complications and partial radiographic success were concurrent with multiple UPDs in the same patient.
There was a relationship between TLE effectiveness expressed as the rate of clinical and procedural success, short-, medium-, and long-term mortality, and the risk of UPDs. Clearly, lower rates of procedural success were noted in patients with UPDs such as Byrd dilator collapse/fracture, lead-on-lead binding, the fracture of the targeted lead, the use of an alternative venous approach, and the loss of a broken lead fragment. These differences were somewhat less pronounced for clinical success; however, they had the same direction of change. This proves a significant impact of these UPDs on the efficiency of TLE. The number of deaths in individual subgroups during the observation period was too small to draw reliable conclusions, but it seems that increased TLE complexity did not influence long-term survival. As previously mentioned the occurrence of multiple UPDs in the same patient significantly reduced the chances of procedural and clinical success (Table 5).    A multivariable regression analysis showed that patient age at first CIED implantation, the number of leads in the heart before TLE, abnormally long lead loops in the atrium or ventricle before TLE, and lead dwell time were independent risk factors for the occurrence of one or multiple UPDs during TLE. Single UPDs were less likely to occur in patients undergoing lead extraction due to CIED infection. Leads on both sides of the chest and the extraction of passive fixation leads were also risk factors for the occurrence of multiple UPDs during TLE ( Table 6).
The Kaplan-Meier survival curves showed that the presence of any UPD during TLE was associated with a better outcome after TLE. This result was probably due to lower patient age during TLE, higher LVEF, a lower Charlson comorbidity index, and a lower percentage of infectious indications for TLE in patients with UPDs during TLE ( Figure 3B).

Results Recapitulation
Blockage in the lead implant vein/subclavian region is caused by too parasternal a puncture of the subclavian vein or occlusion of the vein with dense scar tissue (Figure 4). This problem was present in 297 patients (7.98%). It was not associated with patientrelated risk factors, only slightly associated with the number of leads and the presence of abandoned leads, but it was strongly related to implant duration. In our estimation, this is a UPD caused by a previous operator. Its presence prolonged lead extraction and forced the extractor to use second-line tools i.e., metal sheaths or mechanical rotational threaded sheaths when a standard stepwise and cross-over strategy was utilised. Furthermore, it was weakly related to major complications (MC), lower rates of radiographic success (non-removable lead fragments), and therefore lower rates of procedural success, but not to long-term mortality.
Lead-on-lead binding was the second most common UPD, as it occurred in 242 patients (6.59%). It was not associated with patient-related risk factors, weakly related to the number of leads (and leads being extracted), passive leads and abandoned lead presence and extraction, but significantly related to implant duration. Lead-on-lead binding prolonged lead extraction and forced the extractor to use different tips and tricks to free the targeted leads ( Figure 5). It was strongly related to major complications (MC), more common emergent surgical interventions, lower rates of radiographic success (non-removable lead fragments), and therefore lower rates of procedural success, but not to long-term mortality.

Results Recapitulation
Blockage in the lead implant vein/subclavian region is caused by too parasternal a puncture of the subclavian vein or occlusion of the vein with dense scar tissue (Figure 4). This problem was present in 297 patients (7.98%). It was not associated with patientrelated risk factors, only slightly associated with the number of leads and the presence of abandoned leads, but it was strongly related to implant duration. In our estimation, this  Lead-on-lead binding was the second most common UPD, as it occurred in 242 patients (6.59%). It was not associated with patient-related risk factors, weakly related to the number of leads (and leads being extracted), passive leads and abandoned lead presence and extraction, but significantly related to implant duration. Lead-on-lead binding prolonged lead extraction and forced the extractor to use different tips and tricks to free the targeted leads ( Figure 5). It was strongly related to major complications (MC), more common emergent surgical interventions, lower rates of radiographic success (non- lead described above. Byrd dilator collapse/fracture was described in detail in our previous study [22]. The use of an alternative venous approach was the fifth most frequent UPD, as it was seen in 122 patients (3.28%). We analysed all situations in which the lead implant venous access site was insufficient to complete the procedure ( Figure 6). Similar to the fracture of the targeted lead and Byrd dilator collapse/fracture, this UPD occurred in patients who were younger at system implantation, with better health and a lower Charlson comorbidity index. Its occurrence was significantly related to the number of leads, the presence (and extraction) of abandoned leads, the number of previous CIED-related procedures, the presence of abnormally long lead loops in the heart, the extraction of passive leads, and, perhaps more than the others, lead dwell time. Finally, it was associated with the increased risk of MC and the need for rescue surgical intervention. However, this does not mean that one is a direct consequence of the other. Rather, both problems have the same cause, namely the lead dwell time. . Sometimes it is necessary to use two polypropylene sheaths, which act together as scissors to separate the leads (C,D). Lead removed from the circulatory system, connective tissue remnants (E).
Lead fracture was the third most frequent UPD, as it occurred in 143 patients (3.84%). It was strongly dependent on patient-related risk factors (young age, higher EF, lower Charlson comorbidity index), strongly related to the presence of abandoned leads and the number of earlier CIED-related procedures, and the presence of abnormally long lead loops in the heart, strongly related to implant duration and the number of leads being extracted (all risk factors for the formation of fibrous tissue around the leads). It significantly prolonged the duration of lead extraction and forced the extractor to use second-line tools such as lassos, snares, basket catheters, an alternative venous approach (femoral, jugular, or combined). Furthermore, it was strongly associated with major complications (MC), more common emergent surgical interventions, lower rates of radiographic success (non-removable lead fragments), and therefore lower rates of procedural success, but not with long-term mortality.
Byrd dilator collapse/fracture was the fourth most common UPD, as it occurred in 127 patients (3.41%). The setting and consequences are similar to the fracture of the targeted lead described above. Byrd dilator collapse/fracture was described in detail in our previous study [22].
The use of an alternative venous approach was the fifth most frequent UPD, as it was seen in 122 patients (3.28%). We analysed all situations in which the lead implant venous access site was insufficient to complete the procedure ( Figure 6). Similar to the fracture of the targeted lead and Byrd dilator collapse/fracture, this UPD occurred in patients who were younger at system implantation, with better health and a lower Charlson comorbidity index. Its occurrence was significantly related to the number of leads, the presence (and extraction) of abandoned leads, the number of previous CIED-related procedures, the presence of abnormally long lead loops in the heart, the extraction of passive leads, and, perhaps more than the others, lead dwell time. Finally, it was associated with the increased risk of MC and the need for rescue surgical intervention.

Figure 5.
Lead-on-lead binding. The strong fibrous encapsulation prevents further advancement of the wire to free the lead (A,B). Sometimes it is necessary to use two polypropylene sheaths, which act together as scissors to separate the leads (C,D). Lead removed from the circulatory system, connective tissue remnants (E). Dislodgement of functional lead was the sixth most common UPD which occurred in 34 patients (0.91%). Except the left ventricular and His bundle pacing leads, we did not attach much importance to this UPD, but we entered the event in our database. It was not dependent on patient-related risk factors, slightly related to the number of leads and the presence of abandoned leads, but strongly related to lead dwell time. It slightly prolonged the duration of lead extraction, and it did not influence major complications (MC), the procedural success rate, or long-term mortality.
The loss of a broken lead fragment was the seventh most frequent UPD, which occurred in 22 patients (0.60%). All freed lead fragments were grasped and removed, but this event prolonged the duration of the procedure (Figure 7). It was not dependent on patientrelated risk factors, strongly related to the number of leads, presence, and extraction of abandoned leads, significantly related to lead dwell time, and the extraction of passive leads. It significantly prolonged lead extraction and forced the operator to use secondline tools such as lassos, snares, basket catheters, and the alternative venous approach (femoral, jugular, combined). Its occurrence increased rates of major complications (MC), rescue surgical intervention, and decreased the procedural success rate without influencing long-term mortality. However, this does not mean that one is a direct consequence of the other. Rather, both problems have the same cause, namely the lead dwell time. Figure 6. Use of alternative venous approach. A lead with its proximal end outside the generator pocket (A). If it is not retrievable with the lasso catheter via the superior approach, it has to be pulled down to free the end (B,C). The free end can be grasped again via the superior approach (D). The manoeuvre allows the operator to continue the procedure using conventional techniques (E,F) [32,41].
Dislodgement of functional lead was the sixth most common UPD which occurred in 34 patients (0.91%). Except the left ventricular and His bundle pacing leads, we did not attach much importance to this UPD, but we entered the event in our database. It was not dependent on patient-related risk factors, slightly related to the number of leads and the presence of abandoned leads, but strongly related to lead dwell time. It slightly prolonged the duration of lead extraction, and it did not influence major complications (MC), the procedural success rate, or long-term mortality.
The loss of a broken lead fragment was the seventh most frequent UPD, which occurred in 22 patients (0.60%). All freed lead fragments were grasped and removed, but this event prolonged the duration of the procedure (Figure 7). It was not dependent on patient-related risk factors, strongly related to the number of leads, presence, and extraction of abandoned leads, significantly related to lead dwell time, and the extraction of passive leads. It significantly prolonged lead extraction and forced the operator to use second-line tools such as lassos, snares, basket catheters, and the alternative venous approach (femoral, jugular, combined). Its occurrence increased rates of major complications (MC), rescue surgical intervention, and decreased the procedural success rate without influencing long-term mortality. Figure 6. Use of alternative venous approach. A lead with its proximal end outside the generator pocket (A). If it is not retrievable with the lasso catheter via the superior approach, it has to be pulled down to free the end (B,C). The free end can be grasped again via the superior approach (D). The manoeuvre allows the operator to continue the procedure using conventional techniques (E,F) [32,41].

Discussion
In this study, unexpected procedure difficulties occurred in 687 patients (18.46%), with a single UPD in 454 patients (12.20%) and multiple UPDs in 233 patients (6.26%). All UPDs as specific traps of TLE may prolong the procedure, but should not influence the final effect (clinical and procedural success and major complications). Thus, the proper Figure 7. Loss of broken lead fragment. Image before the start of the TLE (A). The distal lead fragment detached during the lead dissection may flow in the bloodstream into the pulmonary circulation (B). It can usually be captured with a lasso catheter (C) and removed from the circulatory system (D) [38].

Discussion
In this study, unexpected procedure difficulties occurred in 687 patients (18.46%), with a single UPD in 454 patients (12.20%) and multiple UPDs in 233 patients (6.26%). All UPDs as specific traps of TLE may prolong the procedure, but should not influence the final effect (clinical and procedural success and major complications). Thus, the proper identification of procedural complexity may influence the patient management strategy (referral of potentially difficult patients to experienced high-volume centres). Therefore, we tried to predict the occurrence of UPDs using popular calculators of TLE-related risk (SAFeTY-TLE, EROS) [7,8], and scores of procedure complexity (LED score, MB score, Advanced Lead Extraction score, and our recently elaborated LECOM score) [42][43][44][45].
It seems that increased TLE complexity did not influence long-term survival. For mortality after TLE, other factors such as general health status and the necessity of device therapy plays a predominant role [46].
Our study mainly concerned the complexity of the TLE procedure, in particular the occurrence of unexpected difficulties that, on the one hand, prolong the procedure and increase its difficulty, and on the other hand (solved/overcome) cannot be classified as strictly defined complications. The impact of major complications on long-term survival has been the subject of previous [47] and recent reports [48]. They show that the mortality of patients after TLE is determined by the general health condition, but above all by the indications, the most important of which seems to be the extent and advancement of CIED infection. The results of our research correspond well in this aspect with the reports of Gomes et al. [47] and Arabia et al. [48]. A difficult, complex, but uncomplicated TLE procedure does not have a negative impact on long-term survival.

1.
The difficulty of lead extraction refers not only to prolonged procedure time (fluoroscopy time), but sometimes to the necessity of solving a number of lesser-known unexpected procedure difficulties (UPDs).

2.
UPDs are present in nearly one fifth of TLE procedures, and can occur simultaneously.

3.
Unexpected procedure difficulties, which usually force the operator to expand the range of techniques and tools, should be considered as part of the learning and training in lead removal.

Study Limitations
One weakness of this study is the fact that it presents the experience of the same team and the same first operator at three facilities. Therefore, the outcomes of the extraction may not represent the overall safety and efficacy of the transvenous extraction of leads with long implant durations. The data was collected on a systematic and ongoing basis, but was analysed retrospectively. All procedures were performed using all types of mechanical systems, but not laser powered sheaths.