Device Removal, Replacement, and Upgrades

Documentation of Pacemaker Pulse Generator Battery Depletion

Most bradycardia pacemaker pulse generators provide direct or indirect indicators of battery depletion documenting the need for enhanced follow-up. Although different manufacturers use different terminologies for these indicators (e.g., elective replacement indicator [ERI] and elective replacement time [ERT]), these indicators essentially provide warning of incipient battery failure (end of service [EOS] or end of life [EOL]). Additionally, certain nonspecific indicators may alert the physician to early signs of battery wear ( Table 34-2 ).

A change in magnet-activated paced rate remains a common indicator of reduced battery output voltage for pacemakers ( Table 34-3 ). Some pacemaker pulse generator models respond to declining voltages through a gradual reduction in magnet-activated pacing rate; reduced rates indicate the need for enhanced follow-up, with still slower rates indicating ERI or obligatory replacement. Other models demonstrate an abrupt shift in the magnet-activated paced rate at the enhanced follow-up period or at ERI. A demand mode switch from DDD to VVI (a magnet mode switch from DOO to VOO) may occur at the elective replacement time or as an obligate replacement indicator for dual-chamber systems before complete battery failure. Inability to reprogram the device, inaccurate measurement of lead impedances, and an automatically reprogrammed reduction in data storage capabilities to preserve battery life may also occur as the pacemaker pulse generator approaches end of service.

Other secondary parameters suggest gradual battery depletion and can be interrogated remotely. The usual battery impedance in a new pulse generator is less than 1000 ohms. As a pulse generator battery depletes, internal battery impedance increases, providing a secondary indicator of the impending need for replacement ( Fig. 34-1 ). Imminent replacement is not required until more definitive indicators appear, such as a change in magnet-activated pacing rate, display of a specific ERI voltage, or mode switch. The dependence of the patient on the pacing functions of the device needs to be taken into account when timing the generator replacement. For example, a patient in sinus rhythm with complete heart block may not tolerate a mode switch to an asynchronous VVI pacing mode at the elective replacement time. This may lead to the pacemaker syndrome, and it can be quite uncomfortable, especially with pacemaker dependence and the loss of atrioventricular (AV) synchrony.

The physician following the pacemaker needs to be fully aware of the availability and interpretation of telemetered battery depletion curves or battery status screens ( Fig. 34-2 ). Specific devices may also perform internal calculations of anticipated device longevity at current programmed settings. Comprehensive knowledge of the expected ERI or end of service performance of various generators is critical to enable one to anticipate a pacemaker pulse generator's end of service. Ultimately, loss of sensing and pacing capability occurs with battery exhaustion. The rate of battery depletion may accelerate as the device reaches end of service, making timely replacement in dependent patients very important.

Indicators for Replacement of Implantable Cardioverter-Defibrillator Generators

The rate of defibrillator pulse generator depletion depends on the frequency of bradycardia pacing, the frequency of delivery of high-energy shocks to terminate tachyarrhythmias, and data storage. Although the precise longevity of ICD battery systems had traditionally been more difficult to predict than for pacemaker patients, this is no longer the case. Improved battery technology, reduced intrinsic current drain, and excellent ERI indicators and battery depletion curves make predicting end of service for ICD pulse generators as straightforward as for pacemakers.

Various manufacturers have developed relatively straightforward methods to document impending ICD battery depletion. These methods fall into the following three major categories: (1) a measurable reduction in battery voltage that can be acquired through telemetry, (2) an increase in measured charge time to a level that indicates the need for elective replacement, and (3) various device-specific markers that indicate a particular degree of pulse generator depletion. Measurement of charge time to a maximal voltage on the device capacitors is the oldest method for estimating the remaining longevity of an ICD. Most early systems included such markers, though this method of determining elective replacement time requires fully charging the capacitors. This would necessitate an office visit, or one could determine the charge time from a capacitor recharge weeks earlier as the result of programmed automatic capacitor maintenance. Full charges reduce the time to battery replacement. Charge times could also be obtained if the device spontaneously charges and delivers therapy at maximum output, but this process does not always occur opportunely between office visits. More current ICD generators provide an accurate measurement of both internal battery voltage and charge time without performing a full energy charge.

Reduced battery voltage provides a readily useful method of determining generator depletion; the value can be obtained telemetrically. This method is now the most common for marking elective replacement time and end of service for ICDs. Specifically, telemetered voltage can be obtained on initial interrogation of the device and is remotely retrievable in most ICD devices.

Some specific ICD models use battery voltage to produce labels that indicate the beginning, middle, or end of service for pulse generators. These labels can be obtained directly through interrogation of the device in the office. End of service may also be clearly indicated graphically (see Fig. 34-2 ). Other devices show estimated longevity with current usage statistics, similar to that seen for pacemakers.

Regardless of the method used by the device to document end of service, the clinician has the responsibility to increase the frequency of follow-up visits as the unit nears elective replacement time to ensure continued safe function of the system, protection of the patient from tachyarrhythmias, and bradycardia support if required. In the event that the patient receives frequent high-energy shocks just before the anticipated need for device replacement, elective replacement time may occur earlier than was originally planned.

Documentation of Lead Malfunction

A variety of causes of lead malfunction require reoperation, due to primary lead failure or secondarily due to premature battery depletion as a result of excessive current drain (see Table 34-1 ). Primary lead malfunction may be due to outer insulation break, inner insulation break in a bipolar coaxial lead, lead conductor fracture, or lead dislodgement. Battery current drain is increased by (1) high pacing thresholds through the need to increase output voltage, (2) failure to optimize generator output for long-term pacing after lead maturation, or (3) inner insulation break with a resultant low pacing impedance and increased current drain. All of these scenarios can result in premature battery depletion. When reoperation for lead malfunction is performed in these situations with primary lead malfunction, consideration should be given to upgrading or replacing the pulse generator, especially if the battery is old and current drain has been increased.

Lead malfunction can usually be documented by noninvasive telemetric evaluation or remote monitoring. A measured bipolar pacing lead impedance less than 200 ohms suggests an inner insulation break between the two coaxial pacing coils. An outer insulation break may be the result of lead wear or may have been inadvertently caused during initial implantation surgery or reoperation, especially with generator replacement; an inner insulation break between the two coils of a bipolar system occurs most commonly at the subclavian insertion site as the result of crush injury to the lead, especially with leads inserted into the subclavian vein and tied securely in a medial position in patients with a tight clavicular-first rib space. Medial insertion of the lead through the costoclavicular ligament can also cause premature lead wear. Telemetry for lead diagnostics demonstrates low bipolar pacing impedance in patients with an inner lead insulation break. The impedance may vary with manipulation of the pacemaker, which causes intermittent shorting of the two lead conductors ( Fig. 34-3 ).

High pacing lead impedance (generally >1200 ohms) may be the result of lead conductor fracture or an incomplete circuit caused by a loose lead pin-pulse generator connection. Rarely, this may occur with fractures of conductors inside the pulse generator header. The introduction of high-impedance leads makes it essential to compare the impedance at implantation with follow-up impedance measurements and with established acceptable impedance ranges for each lead. Depending on the point of discontinuity, lead impedance may vary with manipulation of the pulse generator or with respiration. Lead conductor fractures may be evident on chest radiographs or fluoroscopy; however, absence of visual evidence does not exclude conductor fracture. A break in the connection of the lead to the generator, or within the lead itself, can produce intermittent loss of energy delivery to the heart, which in turn results in absence of pacemaker spikes. Undersensing, or oversensing due to chatter, may also occur with lead conductor fracture ( Fig. 34-4 ).

The scenario of oversensing may be seen with the recalled high-energy Sprint Fidelis ICD lead system (Medtronic). Electrical chatter due to conductor make-break contacts within the lead can be picked up during in-office interrogation or by remote monitoring. Patients can present with an ICD shock from oversensing due to chatter from the fractured electrode, or they may experience episodic lightheadedness or syncope if they are pacemaker-dependent as pacing is inhibited from output suppression due to the electrical noise. If the episodes of chatter occur in a patient who is not dependent on pacing, or if the episodes are not long enough in duration to trigger tachycardia therapy, no symptoms may be reported.

Lead dislodgement produces intermittent noncapture or failure to sense that may be related to respiration. Pacing thresholds needed to achieve consistent capture may rise significantly. Lead impedance increases or remains unchanged. Fluoroscopy may demonstrate a loose or displaced lead tip but is not always diagnostic.

Special Issues for Implantable Cardioverter-Defibrillator Leads

The ICD lead remains the weak link in the ICD system. Oversensing due to diaphragmatic impulses or extraneous signals may inhibit pacing therapy or lead to inappropriate delivery of “treatment” for presumed ventricular tachyarrhythmias, whereas this actually represents noise sensing. Although fracture and degradation of leads have become less common with transvenous, as opposed to epicardial, ICD lead systems, these problems nevertheless occur with some frequency, necessitating reprogramming or reoperation. Conductor fractures due to specific lead design issues have led to recalls and the need for reoperation in patients who experience a lead fracture ( Fig. 34-5 ). Nevertheless, despite the occurrence of ICD lead conductor fractures that can become so obvious because they can lead to multiple inappropriate shocks in patients with Fidelis leads, insulation breakdown still remains the most common cause of ICD lead failure. Cable exteriorization in the Riata lead series (St. Jude Medical, St. Paul, MN) is a good example of this. Both external and internal abrasions may occur with this recalled lead system. The operator at the time of generator replacement needs to have a full understanding of the implications of lead conductor fracture or insulation breach, the likelihood of this occurring in the future with the specific patient, and what is the best course of action for dealing with these issues in addition to performing a generator replacement.

Depleted battery status is readily evident on routine ICD follow-up both in the office and remotely (as described previously), and integrity of the ICD high-energy coils can be evaluated easily in most current devices. High shocking electrode impedance measurements may indicate lead conductor fracture or a lead-generator interface problem.

Measuring high-energy electrode impedance in early devices required delivery of a shock, either to treat a clinical tachyarrhythmia or as part of a noninvasive testing protocol. As a result, in this early patient population, about 10% of patients undergoing ICD generator replacement due to battery depletion were found to have a previously undetected sensing or defibrillation system failure.

Noninvasive indicators of lead malfunction will nearly always occur before operation; nevertheless, the operator should always test the lead system carefully during a pulse generator replacement procedure and must be prepared to deal with malfunctioning leads at the time of generator change. It is possible that manipulation of a lead in the pocket may uncover a previously undetected conductor fracture. This is especially true for high-energy ICD leads, where movement of the lead in the pocket may uncover microfractures of the conducting cables that become evident after the patient has left the hospital. Newer systems automatically measure high-energy lead impedance at office or remote device interrogation in a manner similar to that used for standard pacing and sensing electrodes. The lower-energy impulses delivered by these devices may be more sensitive to the detection of microfractures of the conductors than would higher-energy shocks, which could “arc” across the gap of a small conductor breach.

Determination of Pulse Generator-Lead Interface Malfunction

Pulse generator-lead interface problems may be grouped into the following three categories: (1) loose, incomplete, or uninsulated connections; (2) reversal of atrial and ventricular leads in the pulse generator connector block (for ICDs, reversal of shocking electrode polarity may also occur); and (3) pulse generator-lead mismatch.

A loose pace/sense lead connection is apparent with noninvasive testing. The device fails to deliver pacing spikes when appropriate, it intermittently fails to sense, and/or it oversenses as a result of chatter due to intermittent contact with the setscrew. Oversensing can result in inappropriately high tracking rates or inhibition of ventricular output. Capture or sensing problems may be exacerbated by manipulation of the device. An uninsulated connection most commonly produces current leakage (an electrical short circuit in the system) that inhibits pacing or sensing. Leakage can occur if a setscrew is not properly insulated or tightened or if sealing rings on the lead header do not prevent body fluid from oozing into the pulse generator connector block around a loosely fitting lead. Leakage around lead header sealing rings may result from a loose lead connection or lead-pulse generator mismatch.

Lead impedance in pulse generator-lead interface problems varies, depending on the specific situation. A loose, unconnected lead that remains in the pulse generator connector block, so that lead header sealing rings prevent fluid from entering, causes a break in the electric circuit and a very high impedance. If fluid enters the pulse generator connector block around a loose lead or at the level of a setscrew and maintains contact with body fluids, the short circuit can produce very low measured impedance. As with lead fractures, impedance can vary with manipulation of the device.

Reversed lead connections (i.e., atrial lead in the ventricular port, and vice versa) should be evident before the patient leaves the implantation laboratory, allowing immediate correction. Some atrial and ventricular leads are marked to enable easy identification, especially preformed passive atrial J leads and straight passive ventricular leads; however, it is not uncommon to place “generic” leads into both chambers, especially straight screw-in leads, which may not be so labeled. Likewise, atrial, left ventricular (LV), and right ventricular (RV) lead pace/sense headers for insertion into ICD ports may all be of the International Standard (IS)-1, so the implanter must exercise care in placing these leads properly into the appropriate locations in the ICD generator connector block. The new IS-4 and DF-4 standard leads have fewer connections to the pulse generator, and they are made so that the pace-sense electrode will not fit into the high energy port. It is also possible, in patients in whom a pacemaker or ICD remains inhibited because of native electrical activity, to see no pacing spikes initially after implantation. To be certain that the pulse generator-lead system functions appropriately immediately after implantation, the device should be programmed to an AV delay shorter than the intrinsic PR interval, and the device should be checked with a programmer after the leads are attached to document appropriate function.

Caution exercised at implantation should avoid reversed leads. We have found the practice of confirming lead serial numbers is a useful way to prevent a reversed lead connection. Additionally, we also always connect the ventricular lead first and ensure pacing in the proper chamber.

Beyond ensuring the presence of adequate and appropriate lead connections to the pulse generator, the battery connector block and leads must be compatible (see later). This issue is less important with the standardization of new lead models used for device upgrades or generator replacements. Incompatibility can, however, result in fluid leakage or loose connections, with resultant loss of pace/sense or shocking capabilities, requiring reoperation.

Detection of Need for Reoperation for Other Reasons

Other indications for pacemaker or ICD generator replacement or lead revision (see Table 34-1 ) generally become apparent through careful patient evaluation. Abrupt pulse generator failure with no antecedent sign of battery depletion is rare but can occur, producing symptoms in pacemaker-dependent patients. In others, abnormal pacing output or rate, lack of pacing output, or inappropriate sensing due to generator malfunction may be detected by remote interrogation or in the office. Of particular importance to patients with ICDs are the possibilities of no output when required to terminate tachyarrhythmias, which may be seen with internal shorts in the Riata lead series due to internal wear of insulation under the high-energy coils. ICD leads can also cause inappropriate shocks due to oversensing of diaphragmatic or lead chatter artifact (see Fig. 34-5 ) and oversensing of extraneous electromagnetic signals, such as surveillance systems or high-voltage generators, that can be sensed as ventricular fibrillation or can inhibit ventricular pacing output. Cellular telephones rarely present substantial interference due to variations in signal frequency.

Development of pacemaker syndrome in patients with implanted ventricular demand (VVI), ventricular rate-responsive (VVIR), or atrial rate-responsive (AAIR) pacemakers presents another indication for device revision. This need should be apparent from history and physical examination, although confirmatory blood pressure or cardiac output measurements may be required. Pacemaker syndrome occurring with an implanted functioning dual-chamber pacemaker with normal lead function must be managed by reprogramming.

Interchangeability of Products From Different Manufacturers

Early ICD leads from different manufacturers were compatible only with ICD pulse generators from the same manufacturer. This is especially evident when replacing generators that used LV-1 leads ( Fig. 34-6 ). The LV-1 lead had no sealing rings and required a particular header with sealing rings for its appropriate use. In later models, manufacturers have adhered to standard header designs for ICDs, initially including IS-1 ports for both atrial and ventricular pace/sense leads and DF-1 ports for high-energy defibrillation lead headers, a 3.2-mm unipolar lead head with sealing rings ( Fig. 34-7 ). The newest agreed on DF-4 standard, which provides four electrical connections combining the functions of a bipolar pace/sense connection with up to two high-voltage connections, should reduce some of the confusion There are two IS-4 connections, the high-energy DF-4 standard for high-voltage (defibrillator) leads and the IS-4 standard for nonhigh-voltage leads (atrial and CS leads and right ventricular pace-sense electrodes). IS-4 and DF-4 connections simplify connections to the pulse generator, except when extra leads for defibrillation are required in individual patients; this would require an adapter to connect to the IS-4 lead ( Fig. 34-8 ).

Though generally uncommon, for procedures involving older, nonstandard ICD connector blocks, however, the operator must be familiar with the existing system of leads and generator in the patient before surgery, and technical support from the manufacturer may be required at the time of the operation. A full range of adapters, or various pulse generator header designs, to mate a replacement generator to the existing leads, must be available. Ensuring tight and proper connections between the generator and the lead, and with any adapters and lead extenders, avoids malfunction and current leak. Although older adapters used an uncured medical adhesive to seal set-screws in the connector block of the device, newer adapters use set-screw seals similar to those found in pacemaker pulse generators.

Special Indications for Replacement of Implantable Cardioverter-Defibrillator Generators

As outlined in Table 34-1 , the ICD generator may need to be replaced for other reasons, as discussed here.

Malfunction of the Generator With or Without Lead Malfunction

Hardware or software errors in the ICD generator, or, more commonly, malfunctioning ICD leads, may result in the need to revise the ICD system. The overall reported incidence of lead-related complications has ranged from 2% to 28%. These complications commonly manifest as inappropriate shocks resulting from oversensing of noise; noise sensing due to chatter caused by a fractured conductor has led to inappropriate shocks and pacing inhibition in some patients implanted with Medtronic Fidelis high-energy electrodes (see Fig. 34-5 ). Alternatively, ineffective shocks caused by shunting of defibrillation energy due to an inner insulator breach may lead to a low impedance route for high energy to be delivered directly back to the pulse generator and can cause pulse generator failure. This can occur with inner insulation breach in the St. Jude Medical Riata lead series, with shunting of current between the superior vena cava (SVC) coil and internally eroded RV conductor cables. Replacement of the ICD pulse generator may be indicated in each of these scenarios, in conjunction with lead revision or extraction, due to premature battery depletion, pulse generator failure, or to avoid another operation in the near future if the battery is already partially depleted.

Upgrading to a Higher-Energy Device or Addition of Hardware for an Inadequate Defibrillation Threshold

An elevated defibrillation threshold detected through noninvasive testing, at the time of elective pulse generator replacement, or through repeatedly unsuccessful shocks clinically delivered in an attempt to terminate ventricular arrhythmias, may require a change in hardware configuration. Options include placing a generator capable of delivering higher defibrillation energy to respond to an elevated defibrillation threshold (DFT) (though most devices are quite capable in this regard today), waveform optimization, repositioning the right ventricular apical (RVA) shocking electrode, or the addition of various other lead systems, including SVC or azygous vein coils, or subcutaneous coils, arrays, or patches to better distribute current around the heart to reduce the DFT, lowering shock electrode impedance for higher current delivery.

Reoperation may be required for antiarrhythmic drug changes that lead to substantial alterations in the DFT, although elimination of the offending medication provides a more straightforward solution. One may also change the drug to one that lowers DFT or add a class III medication such as sotalol or dofetilide to reduce the DFT.

Upgrading to Incorporate Dual-Chamber Pacing Capability

Although the DAVID trial and substudies indicated that there was detriment to RV pacing in ICD patients, the DAVID II trial indicated that atrial pacing to maintain chronotropic competence does not increase heart failure or mortality. This is especially important in patients with congestive heart failure or coronary artery disease who require β-blockers. Upgrading to a dual-chamber device along with generator change can require considerable deliberation when substantial hardware is already in place. Several scenarios may be encountered, each with unique potential solutions.

The patient may have a previously implanted abdominal single-chamber ICD with an epicardial or endocardial lead system. In this situation, the operator has three options: (1) to place an endocardial atrial pacing lead through the subclavian system and tunnel the lead subcutaneously to the abdominal pocket, while upgrading the device to a dual-chamber ICD, or (2) to abandon the abdominal ICD and place an entirely new AV sequential ICD system in the pectoral area ( Videos 34-1 to 34-3 ).

The advantage of long-term stability of thresholds for endocardial pace/sense leads speaks for the approach of adding an endocardial atrial lead and tunneling it to the abdomen, but this also requires that the lead be long enough to tunnel to the abdominal site, making manipulation and positioning of the lead in the atrium more challenging. Alternatively, a lead extender may be attached to a shorter lead, but the adapter adds another weak link. Finally, this approach requires opening both the abdominal pocket and the subclavian site simultaneously, which could raise the risk for cross-infection of the abdominal site, eventually requiring extraction of a very old endocardial or epicardial ICD lead system. We clearly prefer not to have two pockets open at the same time, especially when one involves an epicardial lead system, where infection could be disastrous.

Abandoning the abdominal site altogether is preferable. The new pulse generator and lead system are placed in the pectoral area, preferably on the opposite side to avoid any chance of infecting a preexisting chronic endocardial lead, and DFT testing is performed at implantation; the previous abdominal pocket remains closed during this operation, eliminating the possibility of cross infection between sites. The abdominal generator may be turned off and left in place, or it can be removed some time after implantation of the new system, preferably during a separate procedure.

Lead extraction may also be used in this situation. Removing the preexisting lead, if endocardial, may reduce the risk of long-term vascular complications, and this may be preferable in a young patient, rather than just adding more leads.

Upgrade to a Biventricular, Cardiac Resynchronization System

Upgrade to a biventricular (BiV), cardiac resynchronization system has become one of the most common indications for ICD reoperation, either as a de novo device upgrade or at the time of generator replacement. Upgrade requires generator replacement and insertion of a new CS/LV electrode. We perform venography to ensure patency of the vasculature for the new lead before initial incision. Understanding the venous anatomy is critical to making the correct operative decision. If the subclavian/axillary venous system is occluded, options include lead extraction to produce a conduit or placing the CS lead on the opposite side and tunneling it across the chest to the pulse generator. If a transvenous lead is not an option, an epicardial lead can be placed surgically. Implantation of the new lead and device often requires a pocket revision to accommodate the larger generator. Surgical aspects for upgrade to a BiV system are addressed below.

Risk of Complications Associated With Pulse Generator Change

The risk of complications associated with pulse generator change is relatively low, but significant. A retrospective study from 17 Canadian centers described a complication rate of 8.1% at 3 months. Major complications occurred in 5.8% of patients, and minor complications occurred in 2.3%. In a 12-center Canadian study, there was a 9.1% complication rate, including 5.9% that required reoperation and resulted in two deaths. There was a 3.1% minor complication rate at 12 months for patients undergoing ICD generator replacement due to an advisory indication alone. A prospective, 72-center US study looking at pacemaker and ICD pulse generator replacements showed that major complications occurred in 4.0% of patients at 6 months in patients undergoing generator change alone and in 15.3% of patients at 6 months in patients undergoing generator change with planned transvenous lead implantation at the time of generator change. Major complications were higher with ICD compared with pacemaker generator replacements. Complications were highest in patients who had an upgrade to, or a revision of, a cardiac resynchronization therapy (CRT) device (18.7%).

Magnet Response

The response of a bradycardia pacemaker pulse generator to placement of a magnet can assist in the identification of its manufacturer. Pacemaker pulse generators respond to magnet application by entering a fixed-rate single-chamber or dual-chamber pacing mode corresponding to the type of generator and the programmed mode. Magnet rates vary among manufacturers and may provide a clue as to the origin of the device. To undergo a magnet-activated test, the patient must be connected to an electrocardiographic recorder before the magnet is applied and must remain connected until after the magnet is removed. The first few paced complexes after magnet application may occur at a rate or output other than that seen later in the recording, providing identification data, as well as information regarding the integrity of the pulse generator and lead system (e.g., the delivered pulse width may be reduced during the first few paced complexes to ensure that capture still occurs with an adequate safety margin, the “threshold margin test”). Furthermore, with constant magnet application over the pacemaker, some devices continue to pace at a fixed rate, whereas others cease pacing after a programmed number of intervals. Devices temporarily reprogrammed to a backup mode by electrical interference (e.g., electrocautery during surgery) may exhibit unusual magnet responses.

Radiographic or Fluoroscopic Identification of the Pulse Generator

Pacemaker and ICD pulse generators can be identified from their appearance under radiography. This is the most helpful method of identifying unknown devices. The shape and size of the generator characterizes a particular manufacturer (e.g., square, oval, elongated ellipsoid, round), although pulse generator shape can vary significantly from one device model to another, even when produced by the same manufacturer. Considering that the life span of some pacemaker devices may exceed 12 to 15 years, various shapes and sizes will be encountered.

More specific to identification of the pulse generator are radiopaque markings placed near the connector block that code for manufacturer and device model. These markings appear most clearly under magnified cine-fluoroscopic or radiographic examination when the device is positioned perpendicular to the x-ray beam ( Fig. 34-9 ).

The shape and orientation of internal components, which can often be identified radiographically, provide further clues to the device type, manufacturer, and model. Comparison of these radiographic features (size and shape, identification markings, internal components) with compiled x-ray photographs available from manufacturers facilitates identification of the pulse generator.

Finally, an attempt to interrogate a pulse generator with a cadre of different programmers may identify the pulse generator, unless the battery is so depleted that telemetry communication is not possible.

Radiographic or Fluoroscopic Identification of Leads

Radiographic examination of leads serves two purposes. First, it allows the physician to ascertain the presence of unipolar versus bipolar distal electrodes and the fixation mechanism. Distal active-fixation screws may often be seen directly on radiography, whereas passive-fixation leads have a bulbous tip. Second, radiographic examination may identify lead conductor fractures in which the conductor has clearly separated, leaving a gap, especially with magnified views ( Fig. 34-10 ). Lead information of this sort is important for programming, for selecting an appropriately compatible generator, and for identifying leads for extraction. Fluoroscopy also gives some indication of the degree of fibrosis evident through the real-time motion of the lead and surrounding calcification, information that could be useful if extraction is required.

Radiography of leads involves an examination of the insertion site (e.g., subclavian, axillary, cephalic, jugular, or epicardial), acute bends or fractures in the lead, the location of lead coils beneath the pulse generator in the event that they need to be freed for lead repositioning or extraction, the position of the pulse generator connector block, and a general preview of the character of the connector block-lead interface ( Figs. 34-9 and 34-11 ). The lead should be examined fluoroscopically throughout its course for kinking, fracture, or excessive tension as well as for fixation at the distal tip. A thorough radiographic examination of lead integrity and pulse generator-lead interface before reoperation in pacemaker and ICD patients saves much distress when the pocket is opened.

Radiographic examination will also alert the operator to the presence of any retained or abandoned leads or other hardware that may impact the generator change and/or the need to upgrade a system or replace a malfunctioning lead.

Measuring the Functional Capacity of Implanted Leads

One of the most crucial parts of invasive analysis during reoperation involves measurement of pacing and sensing capabilities in existing leads. Vigorous noninvasive evaluation should give the operator a great deal of information regarding lead viability and functional status, though final verification of lead structural and electrical integrity must be performed at the surgical procedure. If surgery is undertaken for pulse generator replacement, demonstrating viability of existing leads is vital to the appropriate long-term performance of the new battery. Surgery for lead repair or revision itself involves extensive testing of chronic leads to confirm the lead as the source of malfunction, ensure normal operation of other leads, and evaluate new leads for optimal positioning inside the heart.

After the pacemaker or ICD pocket is opened, the pulse generator is disconnected from the leads to enable testing of lead sensing and pacing functions. The lead must be disconnected from the pulse generator cautiously in pacemaker-dependent patients; to avoid prolonged ventricular asystole, the operator should be prepared to connect the lead immediately to a cable attached to a functioning external pacing system. The external device should be activated and be delivering pacing impulses before the ventricular lead is disconnected from the pulse generator in a pacemaker-dependent patient. Alternatively, although it is not usually necessary, a temporary pacing wire may be placed before disconnecting a lead in a pacemaker-dependent patient; such additional instrumentation, however, may raise the risk of infection. Of course, the operator must exercise care not to cut the lead! ( Fig. 34-12 ).

Invasive Testing of the Sensing and Pacing Capabilities of Leads That Have Been Implanted for a Long Time

One of the most important aspects of invasive testing involves measurement of pacing and sensing thresholds in long-term pacemaker and ICD leads. Older generations of nonsteroid-eluting leads show some deterioration in pacing and sensing thresholds during the first 4 to 8 weeks after implantation, then reach a relatively stable level for the long term. This is rarely the case for steroid eluting leads. It is possible, however, for thresholds to continue to increase over time, a change that may now be recognized through remote monitoring. The change in threshold from baseline was greatest with active-fixation, nonsteroid-eluting leads; threshold increases are reduced with passive fixation and steroid elution on all lead types. Noninvasive testing should give the operator some clues as to the usefulness of long-term leads, but invasive testing and inspection confirm their functional utility.

Both atrial and ventricular leads must be tested. If bipolar, they should be evaluated in both unipolar and bipolar configurations. The external pacing analyzer is connected to the lead; pacing and sensing thresholds and lead impedance are determined. The voltage pacing threshold at a fixed pulse width is recorded as the threshold that produces reliable capture. Pacing lead impedance is best determined at an increased output voltage (e.g., 5 V) to ensure accuracy.

Low-voltage pacing thresholds are desirable for long-term leads. This allows programming of the pulse generator output to a reduced level, increasing battery longevity. For chronic leads that have been in place for several years, the operator may decide to accept a pacing threshold (at 0.5-msec pulse width) of up to 2.5 V because this provides 2× pacing safety margin for most pulse generators. However, a pacing threshold of 2.5 V that occurs early after implantation (e.g., within 6 months) may not be acceptable. This suggests excessive early fibrosis around the lead tip and the possibility that exit block and loss of capture will develop in the future if the pacing threshold continues to increase. Care at initial implantation helps ensure lower long-term pacing thresholds and improved sensing capabilities. Higher chronic thresholds may be more acceptable in patients with implanted autothreshold devices that can automatically measure pacing thresholds and adjust output to the lowest required for reliable capture.

Thresholds for sensing likewise tend to increase after lead implantation but less so for steroid-eluting leads. Acceptable measurable intracardiac electrogram amplitudes and slew rates depend on the maximum programmable sensitivity of the new pulse generator. For most systems, P-wave amplitude of 1 mV or more and R-wave amplitude of 3 mV or more constitute minimally acceptable long-term values. Such low amplitudes, however, leave little room for further deterioration in lead function and can lead to problems with sensing. P waves of 1.5 mV or more and R waves of 5 mV or more provide an additional safety margin. If atrial or ventricular ectopy is present, the operator should determine the electrogram amplitude of ectopic complexes to ensure appropriate sensing by the pacemaker. In patients with paroxysmal atrial fibrillation, excellent atrial sensing may be required to detect atrial fibrillation reliably without signal dropout. Higher-amplitude electrograms are required for chronic unipolar leads to allow programming of lower sensitivities to avoid myopotential sensing.

Inadequate sensing or pacing thresholds at the time of generator replacement are indications for placement of a new lead in the affected chamber. This involves either capping an old lead and leaving it in place or removing it. The new lead can usually be placed through the same subclavian or axillary vein, although it is preferable to avoid having too many leads (especially more than four) pass through the same vessel, to reduce the chance of venous occlusion and thrombosis. A single new lead may also be placed through the internal jugular vein, external jugular vein, the contralateral subclavian or axillary vein, or deep into the innominate vein ( Video 34-4 ). The proximal tip can be tunneled to the original pocket to meet a second, functional long-term lead for a dual-chamber pacemaker system if required. Alternatively, an entirely new generator or lead system may be placed on the contralateral side. Lead extraction allows placement of new leads through a conduit produced by lead removal.

If a new system is placed on the contralateral side to avoid vascular overload or due to preexisting occlusion, we prefer to operate on only the new side at the first setting. By opening the pocket on only one side, cross infection is avoided. It is easier to return later to remove an abandoned device that has been inactivated than to risk infection of both operative sites.

Examining the Structural Integrity of Leads and the Lead-Generator Interface

Visual inspection at surgery provides clues to lead integrity. Fluid inside the lead body suggests an outer insulation break but, especially in coradial pacing leads, does not necessarily mandate lead replacement. Fluid may also occur routinely in CS leads with an open lumen. Undue tension on the lead near the fixation site may cause kinking, conductor uncoiling, conductor fracture, or thinning of the electric insulator. A hazy appearance of the insulator surrounding an area of tension or repeated stress is common in older leads. This appearance represents surface erosion of the lead insulator and does not itself imply lead malfunction. The finding should, however, alert the operator to the possibility of lead damage in areas of stress to the insulation. An examination of the suture location ensures that the ligature remains around the suture sleeve, and gentle tension on the lead body ensures its fixation at the venous entry site. Visual inspection of the specific course of a coiled lead in the pocket may be hampered by a significant thickness of overlying capsule scar; fluoroscopy can assist in this regard. Be prepared for any surprise, such as a “twiddled” lead ( Fig. 34-13 ).

Direct examination of the lead connector can assist in the identification of the lead model if not previously known. This is particularly important for lead models that have excessive premature failure rates; such leads in the ventricular position should routinely be replaced in pacemaker-dependent patients.

Venography

Venography is commonly required as part of the device replacement procedure. It plays an important role when insertion of replacement leads into the subclavian vein is difficult, as it can ensure patency of the subclavian and SVC systems, demonstrate points of venous occlusion, or show the course of the axillary venous system for direct access.

Venography is indicated when the subclavian, axillary, or cephalic veins cannot be accessed (to demonstrate their locations), when the veins are accessed but a guidewire cannot be passed into the SVC, and when lead upgrade is required in the presence of longstanding prior lead implant durations. Inability to access the subclavian vein that carries a previously implanted lead suggests either an incorrect needle insertion angle or an occluded subclavian or brachiocephalic venous system (Video 34-5) . Finding an appropriate location to insert the access needle can be facilitated by advancing the needle fluoroscopically in the direction of the chronic electrode under the clavicle, with care not to damage the implanted lead. The vein should be approached with the bevel of the needle facing the implanted lead. If access is not possible, venography may provide better delineation of the course of the axillary or subclavian veins. In this situation, a local injection of radiopaque dye must be injected distal to the veins to be visualized, that is, into the basilic or median cubital vein.

Occasionally, access to the axillary or subclavian vein is possible, but the guidewire will not pass freely to the SVC. If needle placement in the vessel is adequate, failure to pass a wire suggests proximal venous occlusion. Venography demonstrates whether occlusion is indeed present and, if so, its site. Chronic venous occlusion may occur asymptomatically in conjunction with the development of collateral venous circulation around the shoulder. Delineation of the location and length of occlusion indicates to the operator an appropriate needle insertion site for placement of a new lead. It also ensures patency of the SVC. Dye is injected directly into the subclavian or axillary vein through the insertion needle, as this local venogram gives the best opacification. Occlusion of the brachiocephalic system proximal to the junction of the internal jugular vein excludes the ipsilateral jugular system as an alternative site for a new lead. Alternatively, if the subclavian vein is occluded and the internal jugular vein remains patent, a new lead may still be placed on the same side using the jugular approach. Occlusion of the SVC precludes the use of any new endocardial lead placed from a superior site unless leads are extracted to produce a conduit. Although venoplasty is acceptable against preexisting intravascular leads, stents should never be placed into veins without removal of preexisting leads, so as to avoid trapping the leads between a stent and the vein wall ( Fig. 34-14 ).

A persistent SVC (which usually drains into the CS) makes placement of right ventricular endocardial leads difficult or impossible. In some instances of venous occlusion, it may facilitate placement of a lead system. Venography defines the anatomy of the venous system in such a situation, which may be suggested by an unusual intravascular guidewire course. Finally, leakage of venography dye into perivascular tissues or into the pericardial space suggests vessel or cardiac chamber perforation, respectively.

If the patient has had any prior vascular access ipsilateral to the side of the intended new lead implant, we perform venography before incision of the device pocket. Prior vascular access includes subclavian or internal jugular central catheters (both tunneled and nontunneled), tunneled subcutaneous infusion ports, or hemodialysis access ports.

The technique is performed by injection of 10 to 20 mL of radiopaque dye (a 50% dilution generally suffices) followed by a brisk saline flush into a vein peripheral to the occlusion site. Fluoroscopy with permanent storage of cine images is necessary to evaluate flow.

General Guidelines and Techniques

There is no substitute for careful surgical planning in approaching the established pacemaker pocket and gentle handling of the tissues. Perfect hemostasis, avoidance of a tight-fitting pacemaker or ICD pocket, and multilayered incision closure are the basic principles that help prevent future difficulties. These principles are similar to those required at initial implantation (see Box 34-1 ). To avoid induction of ventricular fibrillation, development of fibrosis at the lead tip, and damage to the generator itself, electrocautery must not be used directly over an implanted pulse generator with unipolar leads. This issue has become much less of a problem with the current exclusive implantation of bipolar leads. Electrocautery can be used safely during battery changes as long as the leads are not grounded to the patient, to avoid current shunting directly to the heart. Hemostasis at reoperation can usually be secured with electrocautery or direct ligature. Use of surgical absorbable cellulose or topical thrombin assists in treating persistently oozy pockets. We find topical coagulants most useful if there is damage to the device capsule and the underlying fascia and muscle tissues have been compromised. Clinical judgment should be used in the application of various technical approaches ( Fig. 34-15 ).

Specific Techniques

Local anesthesia is administered most commonly as 1% lidocaine (10-20 mL) infiltrated into the scar line from the previous procedure; additional lidocaine may be given under direct vision once the capsule of the pocket has been defined.

The surgical incision is placed directly over the previous incision. There is no reason that a new incision line needs to be made with device replacement. Healing will proceed appropriately with approximation of the tissues and a tight surgical closure. The skin and subcutaneous tissues are opened with sharp dissection, which is required to penetrate the tough scar tissue and dermal layer. Deeper dissection with Metzenbaum scissors or electrocautery is carried out to delineate the pacemaker capsule. Once the pocket is reached, the fibrous capsule is sharply incised and then extended under direct visualization of the implanted pulse generator and leads. The capsule must be opened far enough to allow extraction of the pulse generator and lead connector assembly without undue force. The posterior capsule should be carefully dissected away from the leads to allow mobility. Access to leads and generator may be facilitated through the use of self-retaining retractors. Extreme care is required throughout the procedure to preserve the integrity of the leads and lead connectors; they must not be punctured with anesthetic needles or cut with blades or scissors. If electrocautery is used to remove tissue from the leads in dissecting them from scar tissue in the posterior capsule, the probe must keep moving over the lead so as to not overheat the lead insulation and thereby damage it. Leads with very thin insulation, including coradial leads and most CS electrodes used with BiV systems, are more prone to heat damage from electrocautery due to their thin outer insulators ( Fig. 34-16 ).

The PEAK PlasmaBlade (Medtronic) applies electrical plasma with a lower peak temperature than conventional electrocautery. It may not only reduce procedure time of device replacement but may also reduce the risk of lead damage when compared with conventional electrocautery. Even with this system, though, it is possible to burn through the outer insulation of a lead; caution is key.

Once the generator is delivered out of the pocket, the leads are disconnected and analyzed. Leads from pacemaker-dependent patients need to be expeditiously reconnected to an external pacemaker (see Fig. 34-12 ). Unipolar pacemaker leads require direct grounding to subcutaneous tissue; the active part of the unipolar generator must remain in contact with the patient before the lead is disconnected. Grounding can best be accomplished through a large surface area ground electrode placed directly into the open pocket. Making contact with this electrode onto the active surface of a unipolar pulse generator allows the generator to be removed safely from the pocket before the lead is disconnected, even in a pacemaker-dependent patient.

After being secured to temporary pacing cables, leads can be completely freed of adhesions up to their entry point into the subclavian vein, if necessary, to examine lead integrity or for extraction. We use low-energy electrocautery sparingly to dissect the leads free of adhesions because the scar tissue could be especially tough and adherent to lead structures. If lead replacement or repair is not necessary, and if the function of previously implanted leads is adequate, dissection of the complete course of each lead may not be necessary, as long as lead connector mobility is sufficient to attach it to a new pulse generator without tension ( Fig. 34-17 ).

Lead malfunction or upgrade from a single-chamber to a DDD pacemaker system or to a BiV system may require placement of an additional lead. If a previously implanted lead is extracted through an occluded vessel using a dilating sheath, a guidewire can usually be inserted into the vascular system through the extraction sheath to maintain a conduit for replacement. In other cases, deep subclavian venipuncture, brachiocephalic cutdown, an internal jugular approach, or tunneling from the other side provide alternative means of inserting a new lead.

After the old pulse generator has been detached from leads and lead integrity and functional status have been ascertained, a new pulse generator can be attached. The principles of generator-lead compatibility must be maintained. Redundant lead coils are placed posterior to the pulse generator, and the pocket is closed with at least three layers of absorbable suture—two subcutaneous and one subcuticular. The operator should be amenable to place additional layers in patients with a lot of fat tissue to prevent wound dehiscence and to prevent subsequent puckering of the skin over the incision line. ICD leads may be tested for defibrillation threshold (if performed) before, or concomitant with, final pocket closure.

At generator replacement or revision, the old capsule needs to be incised or removed. We open the capsule in a medial and inferior direction for two reasons. First, a new device, even if an identical model to the one removed, will never fit perfectly in the original pocket without tension, and second, doing so allows for absorption of fluid and fresh blood flow, which are not possible if the relatively avascular capsule is left intact. This reduces the risk of infection, which is higher at generator replacement than at initial device implantation ( Fig. 34-18 ).

In very thin patients, subpectoral or axillary locations may be required. One can access the subpectoral plane by locating the junction between the sternal and clavicular heads of the pectoralis major muscle and making entry at that point in a bloodless plane, taking care to avoid damage to penetrating neurovascular bundles. Most of these neurovascular bundles are on the posterior aspect of the subpectoral plane. As a secondary approach, the deltopectoral groove can be similarly approached. Alternatively, the muscle fibers of the pectoralis major can be teased apart longitudinally to allow entry to the subpectoral plane, but a dry fascial plane is preferable. Axillary subcutaneous placement of a pacing device is generally avoided because of the possibility of lateral migration of the device, which can be uncomfortable for the patient and can, especially with larger pacemaker devices and ICDs, lead to erosion. When required, however, the axillary location can be entered through direct extension from a subclavian pocket or through a separate axillary incision. The device may be secured in placed in a subpectoral location at that site for more stability by placing a nonabsorbable suture through the suture port in the header and attaching it to firm fascia in the pocket. This suture should have some slack to prevent necrosis. We use a #0 suture for this ( Fig. 34-19 ). The abdominal wall, subcostal, intrathoracic and transiliac positions represent other alternatives for a replacement pulse generator. Nevertheless, a subcutaneous prepectoral approach is appropriate in most patients for both pacemaker and ICD reimplantation.