Cardioverter-Defibrillator Implantation Technique

Updated: Apr 01, 2014
  • Author: Adam S Budzikowski, MD, PhD, FHRS; Chief Editor: Richard A Lange, MD, MBA  more...
  • Print


The skin incision is usually made in the right or left subclavicular area, depending on the patient’s handedness. Generally, it is preferred that the implantable cardioverter-defibrillator (ICD) be inserted on the side opposite the patient’s dominant hand. Other considerations for choosing the implant site include the following:

  • Previous mastectomy or lymph node resection – Such sites should be avoided
  • Recreational activities – In hunters, for example, the site on which the gun butt rests should be avoided for implantation
  • Existing cardiac devices – When an upgrade to a defibrillator is planned, the site where the previous device was implanted is preferred, provided that the venous circulation is sufficiently patent to allow lead placement

The positioning of the incision depends to a large extent on the planned vascular access. If cephalic vein cutdown is planned, an incision in the deltopectoral groove may be preferred for easy visualization of the vein and creation of a subpectoral pocket with minimal bleeding. If axillary vein access is planned, the incision should be guided fluoroscopically by inspection of the position of the first and second ribs.

In patients who have a preexisting cardiac device in place, care should be taken to ensure that the incision both provides easy access to vascular structures and is not too far from the existing system.


Creation of Pocket


For a subcutaneous pocket, an incision 5-7 cm in length is made and carried down to the subcutaneous tissue. Hemostasis is achieved by means of electrocauterization, with care taken to avoid any skin burns that may impede the healing process. Dissection is extended to the prepectoral fascia with electrocauterization, blunt dissection, or both. It should not continue beyond the prepectoral fascia; doing so usually results in bleeding from the pectoralis.

Once the incision is carried down to the prepectoral fascia, the electrocautery is used to create a new plane in the inferior part of the incision with the help of an Army-Navy retractor. The pocket that will accommodate the device is then created with a combination of electrodissection and blunt dissection with the fingers. The pocket should be directed obliquely medially to prevent device migration on the lateral aspect of the chest. The size of the pocket depends primarily on the size of the device that will be used.

Once the pocket has been created and hemostasis achieved, attention is turned toward obtaining vascular access.


For a subpectoral pocket, the author prefers, if possible, to use cephalic vein access for placement of leads; the incision is made in the deltopectoral groove, and the cephalic vein is easily visualized. The incision is carried down to the prepectoral fascia and down to the deltopectoral groove, usually first with the cautery and subsequently with blunt dissection to avoid injury to the cephalic vein. The cephalic vein is then isolated and secured.

Subsequently, the lateral edge of the deltopectoral muscle is lifted and gently separated from the pectoralis minor by using blunt dissection with Metzenbaum scissors. Blunt dissection with fingers may also be used at this point. Careful attention must be paid to hemostasis at this stage of the procedure.

Once the subpectoral pocket has been created, attention is turned toward obtaining vascular access.


Placement of Pacing and Defibrillation Lead(s)

Choice of device type

For the majority of patients, a single-chamber ICD (with only a ventricular lead) is preferred unless sinoatrial (SA) node (ie, sinus node) dysfunction is seen or significant atrioventricular (AV) conduction abnormalities are present. For patients with conduction disease and severe left ventricular dysfunction in which extensive ventricular pacing is expected, one should give careful consideration to placement of a resynchronization device; ventricular pacing has been associated with poor outcomes in this patient population.

Choice of ventricular lead(s)

Several models of defibrillation leads are currently available. The author prefers to use active fixation leads, which allow controlled placement in the intraventricular septum. Passive fixation leads, on the other hand, require a more apical position, thereby increasing the risk for ventricular perforation and the complications associated with it. Several other aspects of the lead design must also be considered, including the sensing circuit and the presence or absence of a proximal defibrillation coil.

The pace-sensed part of the lead either allows sensing from the tip of the lead to the dedicated ring in the immediate vicinity of the distal defibrillation coil or provides a larger sensing vector between the tip of the lead and the distal (right ventricular) defibrillation coil. The defibrillation leads may have 2 defibrillation coils, with the distal coil placed in the right ventricle and the more proximal coil typically extending from the junction of the right atria to the superior vena cava, or it may have only a single distal defibrillation coil.

The single-coil defibrillation system has multiple advantages, including possible extractability in the future, in that there is much less adhesion and scarring in the area of the superior vena cava where the defibrillation coil touches the vein wall. Single-coil defibrillation leads also take up less space in the superior vena cava and hence are easier to place in patients who already have multiple existing pacing leads.

For these reasons, the author prefers to use single-coil leads. This system has tremendous benefits, especially in younger patients, in that it permits the physician to anticipate future needs for placement of new leads, as well as for extraction. This is probably less of an advantage in patients who are older and have a shorter life expectancy, given the expected longevity of the defibrillation system.

Finally, one of the defibrillation lead models comes with a polytetrafluoroethylene (PTFE) coating on the defibrillation coils that may decrease tissue ingrowth [9] and diminish collision artifact in comparison with existing pacing and defibrillation leads. [10]

Placement of lead(s)

The steerability of the lead depends on the shape and thickness of the stylet. At the same time, it must be remembered that insertion of even the softest stylet in the lead dramatically increases the force transmitted via the lead; thus, considerable care must be exercised at all times during lead manipulation.

To cross the tricuspid valve, the stylet must be curved to facilitate passage. Such curving usually does not create any difficulty with traversing the venous system and the right atrium. Should any difficulty arise, a straight stylet may be inserted and the lead advanced to the right atrium. If venous system exhibits significant tortuosity or if multiple leads are already present, a hydrophilic wire should be inserted first, followed by a long peelable sheath, to enable easy maneuvering of the lead.

The lead is then advanced to the right ventricle and subsequently to the right ventricular outflow tract to confirm that it is indeed placed in the right ventricle. In this instance, the lead will be located most anteriorly over the heart silhouette in the right anterior oblique (RAO) projection. This ensures that the lead has not inadvertently crossed a patent foramen ovale (PFO) or, potentially, a ventricular septal defect.

The observation of premature ventricular contractions (PVCs) as the lead is advanced to the ventricle does not confirm right ventricular placement, because placement of the lead in the left ventricle will also result in PVCs.

At this time, the stylet is removed, and another stylet (a soft one with 135° angulation at the distal 2-3 cm) is placed in the lead. Counterclockwise torque is maintained on the stylet as the lead is slowly withdrawn and subsequently allowed to drop to the lower portion of the right ventricle. The lead is then gently advanced, with counterclockwise rotation maintained, to allow septal positioning of the lead.

Once the entire length of the distal defibrillation coil has traversed the tricuspid valve, the lead position is inspected in the left anterior oblique (LAO) projection. In this projection, the tip of the lead should be pointing to the right of the screen, apposed to the intraventricular septum in a perpendicular fashion.

If positioning is unsatisfactory, the lead is withdrawn slightly, the stylet is rotated clockwise or counterclockwise, and the lead is advanced once more. If these steps are not helpful, the lead may not have crossed the central portion of the tricuspid valve and may be caught in the chordal apparatus. In such a case, the lead should be withdrawn to the right atrium and the entire procedure repeated.

Once the fluoroscopic position of the lead is satisfactory, electrical connections should be confirmed (from the positive pole to the sensing ring and from the negative pole to the sensing tip of the connector), and sensing should be checked via the analyzer. Initial sensing should be at least 10 mV; the lead can be repositioned to achieve better sensing if less than 10 mV is seen.

In patients with existing leads, careful attention must be paid to lead placement to make sure that any new lead is remote from the previously placed leads, because there is always a possibility that the collision artifact generated by lead interaction may be detected as ventricular arrhythmia by the device and result in inappropriate shocks. Any lead position on the intraventricular septum from the right ventricular outflow tract to the apical septum may be acceptable; no specific sites have been documented to be particularly advantageous.

Once appropriate sensing has been confirmed, the lead is fixed by extending the fixing screw. After lead fixation, the stylet is pulled to one-third of its length, and the position of the lead is again inspected in the anteroposterior (AP) view to reconfirm that the entire defibrillation coil has passed the tricuspid valve and that there is enough slack left on the lead. At this time, the electrical waveform should be evaluated carefully on an analyzer.

Evaluation of electrical parameters

On the analyzer, it is important to look carefully for ST elevation; patterns of ST elevation have been associated with low dislodgment rates. The lead impedance is checked to verify that it is in the appropriate range (as indicated by the lead manufacturer), and the threshold is checked. Typically, an acceptable impedance range is 300-1100 ohms, depending on the lead manufacturer. A threshold of less than 1 mV is acceptable. The author always uses high-output pacing (10 mV and 1 msec) to make sure that no diaphragmatic capture is seen.

Securing of lead(s)

If the electrical parameters are acceptable, the hemostatic sheath is pulled and split, and a suture sleeve is advanced over the lead to the level of the muscle to provide hemostasis at the access site. If there is significant bleeding from the access site as a result of increased central venous pressure, a purse-string suture may be put around the access site; this usually yields appropriate hemostasis.

Appropriate slack should be provided to allow the lead to move freely without placing any traction on the myocardium. Especially in obese patients, the slack is usually substantial because, in the recumbent position, the diaphragm and the heart are significantly displaced superiorly. In pediatric patients, careful attention should be paid to the amount of slack that is left in the right atrium. If significant additional growth is expected, a large loop may be left in the right atrium to allow the lead to follow the growth of the heart.

The lead is usually secured to the pectoralis with 0 silk sutures, with the initial ties placed on the muscle right below the lead. Several knots should be made for secure placement, but a very tight ligature on the pectoralis should be avoided to minimize the risk of muscle necrosis. If muscle necrosis does occur, the necrotic part of the pectoralis can detach itself from the rest of the muscle, resulting in dislodgment of the lead. Once secured, the lead should be inspected for appropriate slack and adjusted accordingly.

Subsequently, the suture is wrapped around the suture sleeve, and several more knots are securely tied, with constant tension maintained on the silk suture to keep the lead from moving. A minimum of 2 sutures should be applied over the suture sleeve for secure placement of the lead. One should always check to make sure that the lead cannot be moved in the suture sleeve. Once this is achieved, the stylet is fully withdrawn from the lead.

Connection to device

The lead connectors are cleaned with wet and dry gauze to make sure that there is no contamination by blood. If the lead has separate connections for the pacing-sensing portion and for the defibrillation coils (ie, does not have an IS-4 connector), it is critical to ensure that these connections are not switched. Such an error may result in inappropriate therapy from the device in the future. This is particularly critical for defibrillation systems using extended bipolar sensing leads (eg, those from Boston Scientific).

In patients undergoing implantation with a single-coil lead (which has only a distal coil), the channel for the proximal coil must be plugged with the DF-1 connector plug and secured in the device. With the new IS-4 connector, such considerations have become obsolete because there is only one connection for the entire ICD lead.

Some manufactures require that the connector channels be “burped” by inserting the torque wrench in the receptacle before inserting the lead. This should always be done with the IS-1 connectors (for defibrillation coils). If there is more than 1 screw securing the lead in the header, the distal screw must always be tightened first and the lead pulled to ensure good mechanical connection.


Insertion of Generator

Next, an anchor suture is placed in the superior and medial aspect of the pocket. The pocket is washed with an antibiotic solution. Any of several antibiotics can be used; the choice is generally dictated by the facility’s infection control policy. The author typically uses either cephalexin or, in patients with penicillin and cephalosporin allergy, bacitracin or vancomycin. Although there is no direct evidence that this technique is helpful, the orthopedic experience suggests that the mechanical force of the washing may remove at least some contaminants.

Once pocket lavage is completed, the device is placed in the pocket. It is important to place the lead(s) in the bottom of the pocket and then to position the device in such a way that it covers the leads. This protects the lead(s) during any future generator change. With this positioning, the first element of the device that will be encountered in the dissection is the device itself or a header, either of which is far more resistant to mechanical or electrical injury than a lead is.

Careful attention must also be paid to orientation of the device within the pocket. The round part of the device should face the lateral aspect of the patient, and the part where the lead connection is made should face the middle of the chest. In patients who require a subpectoral pocket, an Army-Navy retractor may be necessary to lift the pectoralis initially so as to facilitate device placement.

For devices that are placed subpectorally, interrupted sutures with absorbable material are used to suture the pectoralis major to the deltopectoral muscle. For devices that are placed subcutaneously, an initial layer of 0 absorbable suture (placed in a continuous fashion) is necessary to ensure pocket integrity. When suturing, the author prefers to direct the needle in an inferior-to-superior direction to avoid inadvertently damaging the leads or the device.

Subsequent suture layers are dictated by the body habitus. In patients with a very thin layer of adipose tissue, usually only an intradermal layer is required. In obese patients, another continuous layer of absorbable 2-0 suture material may be necessary. Horizontal suturing should be used. For the intradermal layer, absorbable 4-0 monofilament material is necessary for good apposition of the incision edges.

Once the incision is entirely closed, either cyanoacrylate glue or adhesive strips (Steri-Strips) may be used to cover the incision. If cyanoacrylate glue is used, no further dressing is necessary. If Steri-Strips are used, an occlusive sterile dressing should be applied on top of the incision.


Testing of Defibrillation Threshold

After the lead is secured and connected to the defibrillator and the defibrillator placed in the pocket, but before the pocket is closed, defibrillation threshold (DFT) testing should be done in most patients. Although there is some controversy in the literature over the utility of DFT testing, with some studies showing it to have little clinical relevance, [11] most most practitioners still use it in this setting. [12] The most common reasons for not performing DFT testing include very advanced heart failure, recent heart failure exacerbation, borderline hemodynamic status, and the presence or suspicion of intracardiac thrombi.

After DFT testing is completed, the pocket can be closed.

Steps in testing

Deep sedation is necessary for this part of the procedure to minimize patient discomfort. This is achieved with a combination of midazolam and fentanyl, with propofol, or with etomidate, as appropriate for the patient.

Several techniques may be used for the induction of ventricular fibrillation (VF). The most commonly used techniques include the following:

  • Low-energy shock of 1-1.2 J on the T-wave
  • Brief application of direct current to the myocardium
  • Burst pacing (50- to 60-Hz)

There should be 2 defibrillators (preferably biphasic waveform) in the room, one connected to the patient and one used as a backup. The defibrillator should be charged before VF induction (200 J is usually satisfactory). When VF induction is achieved, the device detects the arrhythmia, charges energy, and defibrillates the heart (with either the first or the second shock). If defibrillation fails, the patient should be immediately defibrillated with an external defibrillator.

Multiple algorithms for DFT testing have been published. Most of these do have some utility for research purposes, but, in practice, it may make sense to follow a published protocol that simplifies the process. [13] The first energy shock is programmed to 14 J, and the distal electrode is programmed to be an anode; the second shock is programmed to be 10 J less than the maximal energy of the device. The 10-J safety margin has been traditionally considered satisfactory, although no randomized clinical data exist to support this view.

If the first shock fails, the device is allowed to redetect the arrhythmia, recharge, and deliver the second shock. If the second shock fails, external defibrillation rescue is necessary. If the initial 14-J shock is found to be effective in terminating the arrhythmia, further DFT is unnecessary, because it is very likely that another 14-J shock would succeed. If the defibrillation threshold is found to be higher than 14 J, another test should be performed at least 5 minutes after the initial test to allow for full hemodynamic recovery.

Management of high defibrillation threshold

If a high defibrillation threshold (< 10-J safety margin) is encountered, there are several maneuvers that can be used to manage it. [14]

The first step is to ensure that no pneumothorax is apparent and to confirm the electrical integrity of the defibrillation system. The next (and easiest) step is to reverse the polarity of the defibrillation vector, particularly if the distal coil has not been programmed to be an anode. If this does not improve the defibrillation threshold, the next step is mechanical or electronic disconnection of the proximal coil of the defibrillation lead (if present).

If this also fails, placement of a subcutaneous coil is usually necessary. [15] With one device (made by St Jude Medical), additional steps can be taken even before polarity switching or disconnection of the proximal coil. Data suggest that adjusting the duration of defibrillation phases in accordance with the shock impedance of the system will improve the DFT.

If all of these maneuvers fail, additional hardware may have to be implanted to improve the DFT. Multiple techniques for additional defibrillation coil placement in the innominate vein, as well as in the coronary sinus, have been described; however, a far easier approach is to place an additional defibrillation coil subcutaneously.

Subcutaneous coil placement

Whenever possible, the author prefers to use a separate incision in the anterior or midaxillary line in the middle of the chest ipsilateral to the device. If this is not possible, the subcutaneous coil may be advanced from the device pocket. Several models are commercially available; in the author’s experience, the Medtronic model 6996SQ works well and is very easy to insert.

Initially, the sheath and the stylet that is driving the sheath should be curved and should be advanced under fluoroscopic guidance over the lateral aspect of the chest posteriorly to the border of the spine. Subsequently, the stylet is removed from the sheath, and another defibrillation coil is passed through the sheath to the subcutaneous tissue in the posterior aspect of the chest.

The sheath is then split, and the coil is secured in much the same way as the lead was secured to the pectoralis fascia. This coil must always be plugged in where the proximal coil plugs into the device generator.

At this time, DFT testing may be repeated. Different configurations may be tried; defibrillation may be accomplished from the subcutaneous coil to the distal coil and device, or the device can be electronically disconnected and defibrillation accomplished between the subcutaneous and distal coils.


Postoperative Care

Appropriate pain medication is necessary after the implantation procedure. Patients in whom the device was placed subpectorally experience significantly more pain than those in whom the device was placed subcutaneously.

Usually, an overnight stay is necessary. The next morning, if the device parameters are within the acceptable range, pain is controlled, and no local complications are present in the pocket, the patient can be safely discharged home. Instructions for appropriate incision care should be provided. If cyanoacrylate glue was used, patients can shower within 24 hours if they refrain from rubbing over the incision site. If Steri-Strips were used, patients should avoid showering until postoperative day 5, at which time the occlusive dressing is removed.

The evidence for postprocedural antibiotic administration is inconclusive and, for the most part, not based on randomized trials. Nevertheless, most practitioners prescribe oral antibiotics for a short period. The author typically uses cephalexin (500 mg every 8 hours for 5 days) for patients who do not have penicillin or cephalosporin allergy and clindamycin (300 mg every 8 hours for 5 days) for patients who are allergic to penicillin.

The patient should also be advised to avoid extreme motions of the arm on the side of the implant for up to 6 weeks. The patient should be seen within a few days after the implantation for a wound check to verify that the wound is healing appropriately.

Anticoagulant therapy is provided as warranted after the procedure. In the BRUISE CONTROL (Bridge or Continue Coumadin for Device Surgery Randomized Controlled) trial, patients at high risk for thromboembolism who remained on uninterrupted warfarin therapy before, during, and after the implantation of a pacemaker or ICD had a significantly lower device-pocket hematoma rate than did similar pacemaker and ICD patients whose antithrombotic treatment was bridged with heparin. [16, 17]

Early in-person follow-up visit improves survival after ICD implantation

An analysis of data from the ICD Registry of the National Cardiovascular Data Registry (NCDR) determined that patients who completed a follow-up clinic visit 2 weeks to 3 months after receiving an ICD, with or without biventricular pacing, had significantly greater survival over the following year than those who did not complete an in-person visit. [18, 19] Although these patients were more likely to be readmitted for cardiovascular causes, these hospital readmissions were largely for arrhythmias rather than heart failure.



Complications of ICD placement include the following:

  • Inadvertent access to the axillary artery rather than the axillary vein
  • AV fistula formation if both vessels (artery and vein) are accessed
  • Thrombosis of the axillary vein or subclavian vein (incidence, 1%-3%) [20]
  • Injury to lung parenchyma or pneumothorax or hemothorax
  • Perforation of any vascular structures, including perforation of the right ventricle and cardiac tamponade
  • Infection of the system, including intravascular hardware and endocarditis (incidence, 1%-7%) [21]
  • Local pocket hematoma
  • Obstruction of the superior vena cava by lead bulk
  • High DFTs and failure to defibrillate
  • Death (incidence, 0.2%)