Perioperative and Periprocedural Management, Electromagnetic Interference, and Cardiac Implantable Electronic Devices

Patients with cardiac implantable electronic devices (CIEDs) frequently undergo other surgical procedures in which there may be interference with operating room (OR) equipment. When the number of patients with CIEDs was small and CIEDs had few features, perioperative and periprocedural management usually involved only placing a magnet on the device. CIEDs had less shielding, and sometimes unpredictable behavior could occur. For instance, phantom device reprogramming was a concern with older CIEDs exposed to electromagnetic interference (EMI), but this is no longer a risk, because programming of modern CIEDs requires complex digital transmission unaffected by EMI.

Features and programming of CIEDs, including the development of implantable cardioverter-defibrillators (ICDs) and cardiac resynchronization therapy (CRT), have become more complicated. There are a multitude of settings and features for both bradycardia and tachycardia that are confusing for even the most experienced practitioners.

Similarly, OR equipment in the past was simpler and unlikely to affect a CIED. However, with the increasing complexity of the OR environment, new sources of EMI are present that can interact in important ways with CIEDs, requiring a comprehensive partnership between the health care teams managing the patient with a CIED and the physicians performing the surgical or interventional procedure. Those without CIED experience will have even more difficulty in choosing the correct perioperative management and settings. It is therefore important that those with appropriate CIED knowledge, either the practitioner who follows the patient's CIED or an appropriate CIED team at the institution performing the procedure, are involved to ensure optimal care is given.

There are little data regarding how to manage devices in the perioperative environment, based on only a few prospective studies and mostly case reports or case series. Consensus and position statements have been written by several organizations, though the recommendations are based mostly on expert opinion. This chapter reviews the available data and recommendations regarding periprocedural and perioperative CIED management.

Considerations for Patients with Cardiac Implantable Electronic Devices Undergoing Surgical OR Invasive Procedures

Clinical Status Affecting Arrhythmic Risk OR Device Function

Most patients with an implanted pacemaker or ICD have underlying structural heart disease and significant intrinsic rhythm abnormities and/or risk of arrhythmias. The perioperative period for these patients has the potential to increase this risk. The emotional and physical stresses of the surgical procedure may result in a higher risk of arrhythmia development. Depending on the type of procedure performed, significant fluid shifts, electrolyte and acid-base alterations, and hemodynamic consequences of anesthetics, the patient may experience ischemia, ventricular function deterioration, and/or ventricular tachyarrhythmias. Furthermore, the use of certain medications can alter CIED function, most notably amiodarone, which can increase ICD defibrillation thresholds. Characteristics of the patient, such as indications for the CIED, type of CIED (pacemaker, ICD or CRT), intrinsic rhythm, and cardiac function, as well as certain features of the CIED, such as rate responsiveness, noise algorithms, and mode switch programming, are all determinants of the risk to the patient and how the CIED responds to the periprocedural environment.

Electromagnetic Interference

EMI exposure in the OR is a significant risk for patients with CIEDs due to operative instruments that may affect CIED function, most notably monopolar electrosurgery (see below). In addition, depending on the type of EMI, strength of signal, duration of cautery application, current path through the body, and proximity to the CIED, the risk may vary. Fortunately, modern-day protective CIED circuitry minimizes the effect of EMI.

Oversensing

Oversensing of a noncardiac signal from monopolar electrosurgery is the most common surgical EMI that will affect CIEDs. The effect is inhibition of pacing or triggering of inappropriate tachyarrhythmia therapies in ICDs. The cutting function of electrosurgery, because it is continuous and of lower voltage, is thought by some to be better filtered out by the CIED, whereas the coagulation function, which is higher in amplitude and of shorter duration, may more likely be oversensed. However, there are little data available regarding this issue.

The consequences of oversensing are minimized by using short bursts of electrosurgery. For instance, if a pacemaker is inhibited for only a few seconds, even in a pacemaker-dependent patient, this will generally have no significant hemodynamic effect. That said, an individual with severe heart failure may not be able to tolerate short episodes of asystole compared with a patient with normal cardiac function. Patients with CRT pacing (most often combined with a defibrillator) pose a unique situation where prolonged electrosurgery inhibition of pacing could result in reduced cardiac function due to emergence of the underlying dyssynchronous contraction of the heart. Some CRT patients have marginal clinical status at baseline, made worse by the stress of surgery, such that the lack of even short periods of CRT pacing may be enough to cause hemodynamic decompensation.

Oversensing on the ventricular channels can be misinterpreted by an ICD as a treatable ventricular tachyarrhythmia. Because of the programmed detection time and time needed to charge the capacitor, which together encompass approximately 10 to 20 seconds, bursts of electrosurgery of no more than 4 to 5 seconds at a time with several seconds in between bursts can usually be considered safe. When electrosurgery is not anticipated to be extensive, this approach can be used without the need to reprogram the CIED or to use a magnet. The risk, however, is that the operator will not be cognizant of keeping the bursts short enough to prevent inappropriate therapies ( Fig. 41-1 ).

Figure 41-1, Inappropriate Implantable Cardioverter-Defibrillator (ICD) Shock in a Patient With a Medtronic ICD, Despite Short Electrosurgery Bursts.

Antitachycardia pacing (ATP) therapy can also be delivered inappropriately. Unlike shocks, which require a charge time before the shock is delivered, ATP can be delivered as soon as detection is completed. ATP is generally not dangerous to the patient, and the quality of electrosurgical noise is such that its detection will most likely occur only in the highest programmed rate zone, where ATP may not necessarily be programmed. Whereas in most instances an ICD shock will not cause harm, the patient may have muscle contractions and “jump” on the table. It is possible, however, that inappropriate therapies (shock or ATP) could result in a true ventricular arrhythmia necessitating appropriate therapy.

Oversensing may also occur on the atrial lead. In this situation, upper rate pacing behavior may occur from ventricular tracking of atrial signals. Alternatively, if the sensed noise meets programmed criteria, mode switching may occur. Though also unlikely to lead to patient harm, it may be confusing to providers, because the pacing mode will switch to VVI or VVIR at the programmed lower rate at a time when the patient, in fact, has underlying sinus rhythm.

Device Reset

The risk of device reset was much higher with older devices than with current CIEDs. Exposure to strong EMI can damage device circuitry or microprocessor timers, leading the CIED to revert to manufacturer-determined nominal settings that provide only basic functions programmed in the read-only memory (ROM) of the device, and those settings in random access memory (RAM) are inactivated ( Table 41-1 ). Sometimes devices can be reprogrammed if a device reset occurs, but Boston Scientific devices enter Safety Core and must be replaced.

TABLE 41-1

Device Reset Settings

Manufacturer	Bradycardia Settings	Tachycardia Settings
Biotronik	SSI or DDI, 70 bpm, 7.5 V at 1.5 msec, unipolar pacing and sensing	VF only at 171 bpm, 12 of 16 intervals to detect, 8 maximum energy shocks
Boston Scientific	SSI, 72.5 bpm, 5 V at 1 msec, unipolar pacing and sensing	VF only at 165 bpm, duration 1 sec, 5 shocks at 41 J
ELA/Sorin	SSI, 70 bpm, 5 V at 0.5 msec, unipolar pacing and sensing	VF only at 190 bpm, 6 of 8 cycles to detect, 4 shocks at 42 J
Medtronic	SSI 65 bpm, 5 V at 0.4 msec, sensing and pacing vector automatically configured	VF only at 188 bpm, 18 of 24 intervals to detect, 6 shocks at 35 J
St. Jude Medical	SSI, 67 bpm, 5 V at 0.6 msec, unipolar pacing and sensing	VF only at 146 bpm detection after 12 intervals, 6 maximum energy shocks

bpm, Beats per minute; DDI, dual (chamber) paced, sensed, inhibited; SSI, single (chamber) paced, sensed, inhibited; VF, ventricular fibrillation.

Damage to the Lead-Tissue Interface

Rarely, electrosurgery in very close proximity to a lead may cause current induction down the lead, resulting in localized scarring at the lead-tissue interface or induction of ventricular tachyarrythmias. The risk is greatest if an insulation breach is present. With appropriate precautions, such as being careful to avoid using electrosurgery directly on the generator or lead, this risk is minimized.

Battery Life

Rarely, in the setting of electrosurgery, devices that have reached their elective replacement indicator (ERI) will transition to end of life (EOL) or, if already at EOL, to nonfunctioning behavior.

Changes in Programming

Phantom programming has rarely been reported and only in older accounts. In these circumstances, CIEDs were reprogrammed while at end of battery life when using a magnet and electrosurgery.

Physical Damage to the Cardiac Implantable Electronic Device Related to Surgical Procedures

There may be changes in CIED sensing and/or pacing thresholds after certain cardiothoracic surgeries, especially left ventricular assist device (LVAD) implantation, which may cause unanticipated lead dislodgement or lead damage ( Fig. 41-2 ) (see Case Study 41-1 ). Placement of an LVAD results in ventricular geometry changes that could alter the lead performance or ICD shocking vectors.

Figure 41-2, Chest Radiograph of Atrial Lead Dislodgement After Left Ventricular Assist Device (LVAD) Implant.

Case Study 41-1

Lead Dislodgement During Lvad Surgery

Age: 63 years
Sex: Female
Occupation: Bus driver
Working diagnosis: Lead dislodgement

History

A 63-year-old woman with a history of nonischemic cardiomyopathy and left ventricular ejection fraction of 13% had a dual-chamber implantable cardioverter-defibrillator (ICD) placed 3 years before admission. Her ICD was placed for primary prevention and was complicated by pericardial effusion and tamponade necessitating pericardiocentesis. One year before admission, she had received an appropriate shock for ventricular fibrillation (VF). She had an atrial lead placed for sick sinus syndrome, though paced only 5% of the time, and has now developed permanent atrial fibrillation.

Over the course of the previous year, she has had four admissions for acute decompensated heart failure. At baseline, she could walk about 100 feet before becoming short of breath and needing to stop. She was seen in the cardiology clinic and reported dyspnea with even minimal exertion, a weight gain of 10 pounds, early satiety, and orthopnea. She was admitted to the hospital for heart failure management and then underwent a left ventricular assist device (LVAD) implant as a bridge to cardiac transplant.

Preoperatively, the ICD was interrogated, and she had adequate sensing and impedances in both leads and a threshold of 1.3 V at 0.4 msec in the ventricular lead. Because of her atrial fibrillation, an atrial threshold could not be checked. The tachyarrhythmia detections were turned off and bradycardia settings were left unchanged at DDD 60-130 beats per minute (bpm), though she had mode-switched to VVI.

The LVAD implant surgery, including time on cardiopulmonary bypass, was unremarkable, although dense pericardial adhesions were observed, likely due to pericarditis from lead perforation at the time of ICD implant. Postoperatively, the patient was brought to the cardiothoracic intensive care unit. Over the next 2 days, she remained in atrial fibrillation, and the primary team reported that telemetry showed frequent premature ventricular contractions (PVCs) versus aberrant beats or possible “pacemaker malfunction.” Any electrocardiogram (ECG) obtained during a prolonged episode showed unusual pacing behavior ( Fig. E41-1 ).

Figure E41-1, Twelve-Lead Electrogram Obtained Postoperatively.

The electrophysiology service was then consulted and asked to interrogate her ICD.

Current Medications

Aspirin 325 mg daily
Furosemide 40 mg intravenously twice daily
Hydralazine 10 mg every 6 hours
Lisinopril 2.5 mg every 12 hours
Metoprolol 50 mg every 8 hours
Sildenafil 10 mg every 8 hours
Warfarin 2.5 mg every night at bedtime

Current Symptoms

The patient reported sternal pain and shortness of breath that were improving since the LVAD implant and extubation from mechanical ventilation. She denied palpitations.

Physical Examination

Blood pressure: 74/57 mm Hg
Heart rate: 87 bpm
General: Alert and oriented female, no apparent distress
Neck: Jugular venous pulse 8 cm
Cardiovascular: Normal mechanical LVAD sounds, point of maximal impulse diffuse and laterally displaced, S1 and S2 were heard but soft
Pulmonary: Clear to auscultation, without cough or wheeze
Extremities: Generalized edema
Skin: Well-healed ICD scar in left pectoral area

ECG (see Fig. E41-1 )

Findings

The ECG demonstrates alternating ventricular paced QRS morphologies.

ICD Interrogation ( Fig. E41-2 )

Figure E41-2, A, Atrial threshold check. The top line is a wireless ECG (proximal coil to can), the second line is the atrial electrogram, the third line is the ventricular electrogram, and the bottom line is the shock electogram (distal coil to can). On the wireless ECG, atrial fibrillation is seen. On the first half of the strip, the atrial EGM does not demonstrate atrial fibrillation, but instead shows a ventricular signal. When an atrial threshold check in AAI mode is completed in the second half of the strip, ventricular capture is seen. B, With a ventricular threshold check in VVI mode, ventricular pacing is seen with a slightly different QRS morphology compared with during the atrial pacing threshold check.

Findings

The ICD settings were DDD from 60 bpm to 130 bpm, paced and sensed AV delays of 400 msec, and a single tachycardia (VF) zone at 190 bpm.

Atrial pacing during threshold testing demonstrates ventricular capture. The P wave measured 6.0 mV compared with 1.4 mV preoperatively. The ventricular pacing threshold was unchanged compared with the preoperative check at 1.3 V at 0.4 msec. Lead impedances were unchanged from pre-LVAD measurements.

Focused Clinical Questions and Discussion Points

Question: What does the ICD interrogation demonstrate?

Discussion: This postoperative interrogation is suggestive of dislodgement of the right atrial lead into the right ventricle. Although the single-lead ECG shows the patient is in atrial fibrillation, the atrial electrogram is not consistent with atrial fibrillation electrograms, but instead is a discrete electrogram occurring simultaneously with the QRS complex and the ventricular electrogram. In addition, atrial pacing creates a QRS complex. Thus, the frequent “PVCs” seen on telemetry were actually due to intermittent ventricular pacing by the RA lead that had dislodged into the ventricle.

The preoperative chest x-ray shows a dual-chamber, dual-coil, left-sided ICD ( Fig. E41-3A ). The postoperative chest x-ray taken immediately after the surgery shows interval placement of a LVAD (HeartWare, Framingham, MA) in the cardiac apex ( Fig. E41-3B ). The right atrial lead has clearly dislodged into the right ventricle. This finding was not noticed by the radiologist or primary team on several chest radiographs.

Figure E41-3, A, Preoperative chest radiograph demonstrating a dual-chamber ICD with leads in the expected position. The arrow shows the atrial lead. B, Postoperative chest radiograph reveals that the atrial lead (arrow) has now dislodged into the ventricle. The LVAD is seen in the left ventricular apex, and a right heart catheter is also present.

Question: How do the bradycardia settings explain the 12-lead ECG?

Discussion: This ECG shows two different paced QRS morphologies due to dislodgement of the atrial lead into the ventricle. The bradycardia settings are DDD from 60 bpm to 130 bpm with a paced AV delay of 400 msec. The first QRS complex is formed by the atrial lead pacing the ventricle. The second QRS complex occurs 400 msec later due to pacing from the ventricular lead, which is appropriately tracking “atrial pacing” even though the atrial lead has displaced into the ventricle. The next paced event occurs after the atrial escape interval of 600 msec determined by the lower rate limit of 60 bpm, because the atrial lead is sensing ventricular activity and not the atrial fibrillation. Again, there is ventricular tracking of this beat, and this pattern continues through the ECG. If the atrial lead was still in the right atrium sensing atrial fibrillation and mode switching, the paced rate would have been 70 bpm.

Question: How did this complication occur?

Discussion: It is unusual to have dislodgement of a permanent pacemaker lead that is 3 years old. ¹ ^, ² Possibilities include dislodgement during freeing of pericardial adhesions, placement of cannulas for cardiopulmonary bypass, or lifting the heart to core out the apex to place the LVAD cannula. It is also possible that the dislodgement occurred at a time of the right heart catheter placed without the use of fluoroscopy just before surgery.

Final Diagnosis

Right atrial lead dislodgement during LVAD implant.

Plan of Action

Options for management include the following:

1.
Repositioning the lead: The lead has dislodged into the right atrium across the tricuspid valve. Repositioning the lead may or may not be feasible, because it is 3 years old and may be scarred into place.
2.
Extracting the lead: As the patient is in permanent atrial fibrillation, she does not require an atrial lead anymore. The lead is only 3 years old, and as it is dislodged, it may be relatively straightforward to extract. That said, there are risks of lead extraction, including cardiac or vascular perforation, pulmonary embolism, hemothorax, and death. ¹
3.
Reprogramming the ICD: Turn off the atrial lead and program the ICD pacing parameters to VVI at 40 bpm. The risk of leaving it in its dislodged position is that the lead may damage the valve and/or lead to increased tricuspid regurgitation. There is also a concern that the exposed screw could lead to cardiac perforation, though this occurrence is quite rare, ² and given her dense pericardial adhesions found at the time of LVAD placement and history of cardiac surgery, it is likely that this risk is low. In addition, there is a risk of the lead causing ventricular ectopy from mechanical irritation.

Intervention

The ICD was reprogrammed to VVI at 40 bpm.

Outcome

Echocardiograms obtained at 7, 30, and 60 days demonstrated moderate tricuspid regurgitation that was unchanged compared with baseline. There was no pericardial effusion on any of the echocardiograms.

Findings

The patient no longer had an indication for atrial pacing, because she was in persistent atrial fibrillation and interrogation of her ICD at 30 days post-LVAD demonstrated no ventricular pacing.

Comments

This case demonstrates lead dislodgement identified following cardiac surgery. Lead dislodgement following either pacemaker or ICD procedures is reported to occur most often within the first 3 months after lead implant, but unusually can happen years after implant. ³ ^, ⁴ This reinforces that surgeons who perform cardiothoracic procedures need to be cognizant of the implantable leads present, no matter when they were implanted, and highlights the need for postoperative interrogation in patients after cardiac manipulation.

The primary concern for this patient was whether the dislodged atrial lead would result in any adverse effects, because there was the programming option of inactivating the atrial lead. If the lead had led to worsening tricuspid regurgitation that was clinically important, then repositioning would have been the better option, though it may have been difficult to maneuver a chronically implanted lead. Balancing the risk of complications from reentering a pocket, including risk of infection, ⁵ ^, ⁶ as well as the fact that she was a bridge to transplant with her LVAD and it was hoped that the duration to transplant would be short, the decision was made to not reenter the pocket or manipulate the lead.

Outcome

At 4 months post-LVAD, the patient received a cardiac transplant and her ICD system was removed completely.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here