Introduction

Cardiac perforation is an uncommon but serious complication associated with pacemaker and implantable cardioverter defibrillator (ICD) implantation. The reported incidence of cardiac device lead perforation ranges from 0.1% to 0.8% for pacemaker leads and from 0.6% to 5.2% for ICD leads (1). However, the true prevalence is probably underestimated, given that the diagnosis can be missed in patients with mild or no symptoms (2).

Patients with acute lead perforation (occurring during the implantation procedure or shortly afterward) typically present with hemodynamically significant pericardial effusion, necessitating urgent pericardiocentesis and lead revision. On the other hand, the clinical presentation in patients with subacute or delayed lead–related perforation (occurring 1 to 30 days or >30 days post-implantation, respectively) is highly variable, and many patients remain asymptomatic.

The optimal management of patients with subacute or delayed lead perforation remains uncertain. Various small cohort studies have reported different open surgical and percutaneous techniques and outcomes for lead management; nonetheless, there is a lack of consensus as to the preferred management strategy (3–8). The primary aim of this study was to determine the feasibility and safety of the percutaneous management of subacute and delayed lead perforation.

Methods

This is a single-center retrospective study of all consecutive patients who were diagnosed with subacute or delayed cardiac perforation by a pacemaker or ICD lead and referred for perforating lead revision at the Prairie Heart Institute from January 2007 to February 2020. The study was approved by the Springfield Committee for Research Involving Human Subjects.

For each case of cardiac perforation, detailed demographic data were recorded, including presenting symptoms, electrophysiological (EP) parameters at device interrogation, and diagnostic testing, including pre-procedure chest radiography, transthoracic echocardiography (TTE), and computed tomography (CT).

Subacute lead perforation was defined as perforation diagnosed within 1 to 30 days after implantation and after hospital discharge, with normal electrical parameters, normal chest radiography findings, and no clinical signs of perforation in the first 24 h after implantation. Delayed perforation was defined as that occurring >30 days after implantation. Acute lead perforations, which occurred during or shortly after implantation (<24 h), were excluded.

Cardiac perforation was suspected when patients presented with the following: 1) chest pain, dyspnea, or extracardiac muscle stimulation; 2) pericardial effusion; 3) significantly altered electrical parameters on device interrogation; and 4) evidence of lead migration on imaging studies. Chest x-ray films and TTE were performed in all patients at the time of presentation, and CT was performed as needed to confirm the diagnosis of lead perforation.

The perforating lead revision procedure was performed in either an EP or a hybrid laboratory, with surgical backup support. The method of lead extraction and any associated complications were assessed with reference to the definitions provided by the Heart Rhythm Society consensus document (7). TTE was systematically performed at the end of the procedure to rule out pericardial effusion and repeated within 24 h afterward. Chest x-ray and device interrogation were performed before hospital discharge. Routine follow-up included a first visit to the outpatient clinic 7 to 14 days post-discharge and a second visit 3 to 4 months afterward.

Results

During the study period, 54 patients with subacute or delayed cardiac perforation were identified (35 females; mean age 75.5 ± 9.7 years; range 46 to 95 years). Demographic characteristics are summarized in Table 1. Cardiac device and lead data are summarized in Table 2.

All patients had records including baseline device interrogations and chest radiography within 24 h after the initial implantation procedure that confirmed the presentation and symptom onset beyond the first day post-implant. Lead perforation was subacute in 29 patients (53.7%) and delayed in 25 patients (46.3%). The average time from lead implantation to first clinical presentation of perforation was 60.8 ± 89.1 days (median: 24.5 days; range 2 to 412 days). Cardiac perforation was related to a pacemaker lead in 36 patients and an ICD lead in 18 patients. The perforating lead was an active-fixation model in 51 cases and a passive-fixation model in 3 cases. The culprit lead was originally placed in the right ventricular (RV) apex in 41 patients, in the RV outflow region in 11, and in the right atrial (RA) free wall in 2.

The main symptoms were sharp chest pain (n = 17); dyspnea related to pericardial effusion (n = 8) or hemothorax (n = 2); diaphragmatic stimulation (n = 7); abdominal pain (n = 2); chest wall bruising (n = 3); and symptoms related to bradyarrhythmias caused by pacing malfunction (n = 11) such as dizziness, fatigue, or syncope (Central Illustration). Twenty-three patients (42.6%) were asymptomatic, in whom lead perforation was discovered incidentally on radiographic imaging, suggesting lead migration (chest x-ray films in 3 patients and CT in 4 patients) or anomalous electrical data on routine outpatient device interrogation (n = 16) (Table 3).

The diagnosis of lead perforation was confirmed by device interrogation, chest x-ray films, TTE, or CT. Chest x-ray films and TTE were obtained in all patients (Table 3). Chest x-ray films were suggestive of lead migration (compared with baseline post-implantation x-ray films) in 31 patients (57.4%) (Figure 1). TTE revealed circumferential pericardial effusion in 8 patients (14.8%); the perforating lead was originally implanted in the RV outflow region in 4 patients (all pacemaker leads) and in the RV apex in 4 patients (3 pacemaker and 1 ICD leads) (Figure 2). All patients with pericardial effusion presented with subacute lead perforation. None of those with delayed perforation experienced pericardial effusion.

Chest CT was performed in 25 patients (46.3%). In 21 patients, the CT was obtained after the finding of anomalous lead electrical parameters on device interrogation and nondiagnostic chest x-ray films; in all 21 patients, CT findings were diagnostic for lead perforation (Figure 2). In 4 patients, chest CT was obtained for unrelated indications, and lead perforation was found incidentally. Lead tip migration beyond the pericardium was confirmed on chest x-ray films or CT in 19 patients (35.2%). Of those, the tips of the perforating leads were localized within or below the left hemidiaphragm in 9 patients, inside the pleural space in 2 patients, and in the anterior chest wall (within the intercostal or the pectoralis muscles) in 8 patients.

At the time of diagnosis, all 54 patients demonstrated altered electrical parameters compared with those at implantation (Central Illustration). Pacing impedance of the perforating lead increased to >2,000 Ω in 13 patients and decreased to <250 Ω in 6 patients. All 6 patients with decreased lead impedance also had pericardial effusions. A significant decrease (by >50%) in electrogram amplitude was observed in 27 patients. Additionally, bipolar pacing threshold increased by >1 V in 29 patients, including 21 patients with complete loss of capture at maximal output stimulation. In 12 patients with no significant (i.e., <1 V) changes in bipolar pacing thresholds, unipolar pacing threshold testing from the distal electrode of the perforating lead revealed significantly higher (>1 V) capture thresholds compared to bipolar capture thresholds. Diaphragmatic stimulation during high-output pacing from the culprit lead was observed in 18 patients. Furthermore, chest pain could be reproduced in 15 patients with high-output pacing from the perforating lead. Of note, 16 patients exhibited noise at stored intracardiac electrograms. No inappropriate shocks were recorded in patients with ICD.

In all patients, the leads were removed or repositioned percutaneously through subclavian vein access under fluoroscopic guidance. The procedure was performed in either an EP (n = 41) or a hybrid laboratory, with surgical backup support. Transesophageal echocardiography was used during the procedure in 12 patients, and intracardiac echocardiography was used in 7 patients for monitoring of new or worsening pericardial effusion (Figure 1). In patients presenting with pericardial effusion, pericardiocentesis was performed before the lead revision procedure, with a pericardial drain left in place until after the procedure (Figure 2). Similarly, chest tubes were placed in the 2 patients with hemothorax before the procedure. Pericardial and pleural fluids were bloody in all cases, with negative culture and cytologic findings.

In 29 patients, the perforating lead was manually retracted and then repositioned to a different intracardiac location without removing the lead from the patient’s body. When the perforating lead could not be easily manipulated, a new lead was implanted, and the perforating lead was completely extracted by manual traction (n = 21) with or without using a locking stylet or was extracted using a laser-powered sheath (n = 4). All 4 leads requiring laser extraction were dual-coil ICD leads. None of the patients required surgical intervention.

All patients were followed up for a minimum of 3 months, (mean follow-up time: 32.3 months, range 3 to 126 months). Two patients developed moderate-sized pocket hematomas that resolved completely with conservative management. All 8 patients with pericardial drains developed chest discomfort consistent with pericarditis in the first few days after the procedure and were treated anti-inflammatory drugs. Post-procedural TTE did not show new pericardial effusion in any patients. No device-related infections or other periprocedural complications were observed.

Discussion

Main findings

The main study findings were that: 1) subacute and delayed lead perforation can occur with different lead profiles and intracardiac locations, but RV apical implants were the most common (75.9%); 2) the clinical presentation of subacute and delayed lead perforation is highly variable, and a large proportion (42.6%) of patients were asymptomatic; 3) all patients exhibited altered electrical parameters on device interrogation; and 4) all of our patients with subacute or delayed lead perforation were successfully treated with the percutaneous lead management approach with no major periprocedural complications, and none required cardiothoracic surgery.

Incidence and mechanism

The incidence of lead-related cardiac perforation post-pacemaker and post-ICD implantation is small but not negligible. In a recent meta-analysis comprising 60,744 patients, the incidence of lead perforation ranged from 0% to 6.4% (mean: 0.82%; median: 0.40%) (9). Late lead perforation is reported in 0.1% to 0.8% of pacemakers and 0.6% to 5.2% of ICD implantations (1). However, the true prevalence is probably underestimated, given the fact that patients are frequently asymptomatic (2).

Perforations may occur acutely at lead insertion or, alternatively, be delayed for days, months, or even years. According to the timing of presentation, lead perforations are classified as acute (manifesting within 24 h of the implant procedure), subacute (detected within 30 days of implantation), or delayed (diagnosed later than 30 days post-procedure). In our cohort, lead perforation was subacute in 29 (53.7%) patients and delayed in 25 (46.3%) patients.

Several studies described certain factors associated with increased risk of lead perforation, including lead characteristics (e.g., lead stiffness, tip diameter, active-fixation mechanism), endocardial location (RV apex), low operator experience, and patient-specific characteristics (advanced age, female sex, low body weight, and use of systemic steroids) (3–5). In our patient cohort, two-thirds of the culprit leads were pacemaker leads, and 65% of patients were elderly women. Consistent with previous reports, the vast majority of perforating leads were active-fixation leads, and 76% were originally implanted in the RV apex. However, as shown in our study, cardiac perforation also can be caused by leads implanted in the RV outflow region in a non-negligible proportion of cases, as well as in the RA.

The precise mechanism of subacute and delayed lead perforations has not been fully elucidated. Although acute perforations are likely caused by forceful advancement of the lead tip against the myocardial wall or overextension of the fixation helix, delayed presentations suggest a more gradual process, with progressive changes in tissue integrity or the result of chronic mechanical action at the lead tip. One proposed mechanism involves advancement of the lead tip during cardiac contraction, dissecting between muscle fibers (10). Alternatively, excessive tension applied on the lead so that its tip is stabilized in a fixed place, not following the heart beats, can potentially cause gradual substrate corrosion as a result of the continuous pressure on the myocardium (3). These mechanisms explain the low frequency of cardiac tamponade with delayed perforations; the gradual process allows the self-sealing properties of the myocardial wall by muscle contraction, fibrosis, or by the lead itself to reduce the risk of rapidly accumulating pericardial effusion (8). In our cohort, only 8 patients (14.8%) presented with pericardial effusion, and all of those patients had subacute lead perforation. None of those with delayed perforation experienced pericardial effusion.

Clinical presentation and diagnosis

The clinical presentation of lead perforation is highly variable. Although patients with acute lead perforation typically present with hemodynamically significant pericardial effusion, those with subacute or delayed perforations can present with a variety of symptoms, and many remain asymptomatic. In fact, in more than 40% of our patients, lead perforation was discovered incidentally on radiographic imaging suggesting lead migration or anomalous electrical data on routine outpatient device interrogation. Therefore, a high index of suspicion should be maintained to identify this potentially serious complication (11).

Chest pain, abdominal pain, extracardiac muscle stimulation, dyspnea, and bradycardia-related symptoms in pacing-dependent patients were experienced by more than one-half of the patients in our study. These symptoms should prompt further investigation, including detailed device interrogation, chest x-ray films, and TTE.

Altered lead electrical parameters at device interrogation in the culprit lead compared with immediate post-implantation values were observed in all of our patients and often were the first clue to the diagnosis of cardiac perforation. Therefore, even a small increase in pacing or sensing thresholds should raise suspicion of a change in lead position. It is important to note that the proximal ring electrode may still be in contact with the myocardium in the setting of a small perforation, resulting in preserved pacing function (anodal capture in bipolar configuration). Therefore, when lead perforation is suspected and bipolar pacing capture thresholds are not dramatically changed, unipolar capture threshold testing from the tip of the culprit lead can unmask pacing malfunction. These finding are in concordance with observations in other studies (12).

The initial investigations for lead perforation include chest radiography and TTE. In some cases, fluoroscopy can help confirm the diagnosis by demonstrating that the lead tip does not follow the heart wall motion. If the results of those tests are inconclusive, chest CT can be valuable to confirm the diagnosis. CT also can determine the extent of lead perforation and help with planning the management approach (8). Recent studies showed that the accuracy of CT imaging for the diagnosis of cardiac perforation was superior to both TTE and chest x-ray films (5,13). Our findings confirmed these previous observations. In our study, chest CT was performed in 25 patients. In 4 of those patients, the CT was obtained for unrelated indications, and lead perforation was found incidentally. In 21 patients, the CT was obtained after the finding of anomalous lead electrical parameters on device interrogation and nondiagnostic chest x-ray films; in all 21 patients, CT findings were diagnostic for lead perforation.

Management

The optimal treatment of patients with lead perforation remains uncertain. Once cardiac perforation is recognized, lead revision is generally recommended to avoid further lead migration (14). Various transvenous and surgical methods for the management of lead perforation have been reported. Recently, several studies proposed percutaneous lead revision with surgical backup as an alternative to open surgical procedures, although the data are very limited (6,8,11). The findings in our study are consistent with those of earlier studies; all of our patients were successfully treated with percutaneous lead repositioning or replacement as the initial management strategy, with no major periprocedural complications, and none required rescue cardiothoracic surgery. Lead repositioning (without lead removal) was not always feasible because of the presence of fibrotic adhesions, which is often related to the length of time elapsed since implantation. Lead extraction was performed predominantly by using manual traction under fluoroscopy guidance, with or without a locking stylet. The use of laser-powered sheaths was needed in a minority of cases, which is expected, given that all leads were implanted within <2 years.

According to the 2017 Heart Rhythm Society expert consensus statement (7), lead extraction should be considered if lead perforation causes significant symptoms, pericardial bleeding, or device malfunction. However, the optimal approach in asymptomatic lead perforation remains debated. Some investigators proposed a conservative management approach for perforating leads with no consequent symptoms or clinically significant lead malfunction (7,12). On the other hand, a recent study showed that conservative management of lead perforation is associated with increased complications (resulting from further lead migration) compared with early lead revision (14). In the present study, asymptomatic patients with lead perforation (n = 23) were treated uneventfully with invasive lead revision.

An important potential complication of percutaneous management of lead perforation is acute life-threatening pericardial bleeding developing shortly after retraction of the perforating lead tip, which has been described in previous reports (15,16). Therefore, an initial open surgical approach was advocated by some investigators and was the preferred approach in a consensus statement endorsed by the American Heart Association (11,15,17–20). However, in our experience and that of others, neither atrial nor ventricular perforation led to new-onset or worsening pericardial effusion, likely because myocardial self-sealing properties at the site of perforation (5,12). Therefore, surgical extraction is not necessary in the majority of cases. Additionally, a previous report proposed the insertion of a pericardial drain prophylactically, before the extraction procedure, in selected high-risk patients without pre-existing pericardial effusion to help reduce the need for any emergent pericardiocentesis; however, the benefit of this approach remains unproven (12,16). We have used this approach only in those patients with pre-procedural pericardial effusion; no patients without pericardial drain catheters developed pericardial effusions. Nonetheless, meticulous intraprocedural monitoring is essential for prompt recognition of the development of cardiac tamponade and the need of urgent pericardiocentesis (12). Use of transesophageal or intracardiac echocardiography can potentially add to procedural safety.

Surgical repair also has been proposed as the initial management approach when the perforating lead perforation has migrated into another cavity, such as the pleura, peritoneal cavity, or visceral organs. Nevertheless, in a recent study, the percutaneous approach was successfully used for lead perforations beyond the pericardial space (3,12). Our results are in concordance with these observations. Nineteen patients in our study had confirmed lead migration beyond the pericardium, and they all were treated successfully without surgical intervention.

It is important to emphasize that these procedures are high risk and should be performed in experienced lead management centers with multidisciplinary teams and with the support of surgical backup. Detailed pre-procedural imaging to accurately define lead tip position and relationship to neighboring structures is critical. Careful hemodynamic and, in select patients, echocardiographic monitoring is important to ensure an optimal outcome. The choice between a well-equipped EP laboratory versus an operating room often is dictated by the individual case pre-procedural risk profile (12).

Study limitations

First, this was a retrospective, single-center study, and inherent limitations associated with this type of design should be taken into consideration. Second, the number of participants included in our study is relatively small, which is anticipated given the low prevalence of lead perforation; nonetheless, our study cohort is the largest reported so far.

Our institution is a tertiary referral center for complex lead management, and many patients in our cohort were transferred from other centers where they had originally received their implants. Therefore, the details regarding implantation technique and lead selection were not available. Moreover, the incidence of lead perforation could not be estimated because the total volume of device implants was not available. Furthermore, given the lack of comparison with a control group, no conclusions can be drawn on risk factors for lead perforations (such as RV apical versus outflow tract placement or the use of active- versus passive-fixation leads) or on the diagnostic value of lead parameter changes. Of note, although all patients underwent chest radiography and TTE, not every patient underwent CT evaluation, and there was heterogeneity in the diagnostic protocols used. However, the management protocol was consistent: the percutaneous lead approach was used as the initial management strategy for all patients referred to our institution for lead perforation.

Conclusions

Subacute and delayed cardiac perforation is a rare but serious complication of pacemaker and ICD implantation. Our study shows that lead perforation can occur with different lead profiles and intracardiac locations and can exhibit a widely variable clinical presentation. A large proportion of patients can remain asymptomatic and only exhibit altered electrical parameters on device interrogation; hence, a high index of suspicion is required to establish the diagnosis. Our data, the largest case series to date, show that patients with subacute or delayed lead perforation can be successfully treated with percutaneous lead repositioning or replacement as the initial management approach, without the need for cardiothoracic surgery. Importantly, these procedures are high risk and should be performed in experienced lead management centers with multidisciplinary teams and with the support of surgical backup.

Footnotes