Permanent Pacemaker and Implantable Cardioverter-Defibrillator Implantation in Adults

Physician/Surgeon

Although training, competency, and credentialing for CIED implantation should all be substantially linked, even now, as in the past, a number of training pathways and guidelines exist and are both proposed and supported by professional organizations whose constituencies vary significantly. These organizations and their constituencies, as well as hospital-credentialing entities, are motivated by a variety of factors. The enunciation of training requirements, evaluations for competencies, and credentialing criteria necessarily are influenced by many factors. Most importantly, all of these factors aside, CIED implantation training, competency, and credentialing should be governed by best patient care. With historical certainty, regardless of whether CIED implantation has been performed by cardiothoracic surgeons, device-focused adult/pediatric cardiologists, or adult/pediatric electrophysiologists, background training and competency has been variable. It is beyond the scope of the responsibilities of the authors of this chapter to be overly prescriptive in this regard. Some generalities of competencies, however, bear discussion.

Inherent differences in competencies exist between surgeons and adult/pediatric cardiologists and electrophysiologists. By training and experience, surgical aspects of CIED implantation are generally less challenging for those trained as surgeons in comparison with those trained as cardio­logists or electrophysiologists. On the other hand, cardiologists and electrophysiologists have substantially more training and inherent competencies in catheter and lead manipulation within the vascular system and heart, as well as imaging, than do most surgeons.

A reasonable set of training requirements for adult cardiologists and electrophysiologists is identified in the 2008 Adult Cardiovascular Medicine Core Cardiology Training (COCATS) Taskforce 6 report. These requirements, in essence, require 300 CIED interrogations/programming sessions, 75 initial CIED implants (including 25 ICDs, 25 dual-chamber devices, and 25 CRT devices), and 30 CIED revisions/replacements. These recommendations, rather than those of surgical and pediatric organizations, are mentioned here because the vast majority of CIED procedures throughout the world are being performed by cardiologists and electrophysiologists who treat adult patients. Challenges exist for pediatric CIED implantation and for surgeons who want to develop expertise in this area, largely because of limited numbers of procedures in the former and time commitment in the latter. Although hospital-credentialing entities increasingly use training recommendations of specialty boards and professional organizations, substantial variability in the credentialing requirements continue to exist. Evolution of implantation training and competency, as well as credentialing, will continue to evolve with the evolution of CIED technology. New recommendations for training are anticipated in an American College of Cardiology/American Heart Association/Heart Rhythm Society Electrophysiology Advanced Training Statement in late 2015.

Although the linking of interrogations and programming sessions into the criteria for implantation competency could seem misplaced, it is important to recognize that substantial complications can occur if implanters do not understand both basic and sophisticated electrical performance issues with implanted devices. Only by being involved in interrogations and programming, including routine and troubleshooting sessions, will implanters best be able to anticipate and prevent problems with implantation decisions and techniques. Similarly, it should be pointed out that in gaining experience with CIED implantation, a variety of anatomic and cosmetic issues must be encountered for achieving true competency. In this regard, implanters should understand the variety of venous access options, retromuscular and premuscular device implantation, and grasp techniques available for tunneling procedures. These are discussed further in this chapter.

Support Personnel

A detailed discussion of many of the nuances of procedural support personnel is contained in a Heart Rhythm Society Expert Consensus Statement. Important issues of training, certification, and credentialing for the variety of potentially helpful personnel must be attended to for medical, legal, and other regulatory reasons.

Although availability of anesthesiologists and/or certified anesthetists is variable, there is no doubt that in certain situations such as airway, respiratory, and even circulatory instability, such personnel can be extremely helpful. Registered nurses (RNs), “scrub” assistants, radiology technologists (RTs), cardiovascular technologists (CVTs), and industry-employed allied professionals (IEAPs) are commonly involved in support of CIED implantation. Advanced practice nurses (APNs) and physician associates (PAs) are frequently involved in preprocedural and postprocedural care of CIED patients but less commonly involved in the implant procedures directly. In general, ideally, all personnel involved in CIED implantation procedures (with the exception of IEAPs in attendance to provide the devices, leads, implantation tools, and device and lead testing) should have ACLS training and certification. Those involved in pediatric/CIED implantation ideally should have PALS training and certification. At a minimum, the RNs involved in sedation, blood pressure, oximetry, capnography, heart rhythm monitoring, and external defibrillation should be thusly trained and certified. Just as with implanting physicians, support personnel experienced in cardiovascular anatomy and physiology and with substantial numbers of patients with various degrees of cardiovascular illness are invaluable. Although cross-training of implant personnel can be very helpful, typical responsibilities for support personnel in implant procedures include (optional cross training designated below within parentheses):

• RN (CRNA) for sedation, monitoring of airway, BP, oximetry, capnography, heart rhythm, and defibrillation

• “Scrub” assistant (RN, CVT, RT)

• Circulating assistant (RN, CVT, RT) for documentation and for provision of needed supplies and other support

• IEAP (RN, CVT, RT) for provision of implantable hardware, tools, and testing

All personnel must be supervised during procedures by the implanting physician, and all, including IEAPs, must have the required hospital/facility credentialing.

Preoperative Orders

The preoperative orders for CIED implantation are generally simple. Although governed by hospital policies and the plans for sedation/anesthesia, in general, the patient fasts for at least 6 hours before the procedure. If the implantation is an ambulatory procedure, the patient reports to the hospital on the day of the procedure, with enough time to obtain the necessary preoperative testing, generally 2 hours. The preoperative procedures consist of posteroanterior (PA) and lateral chest radiographs; an electrocardiogram (ECG); a complete blood count (CBC); prothrombin time (PT); partial thromboplastin time (PTT); and measurements of serum electrolytes, blood urea nitrogen (BUN), and serum creatinine. Because the patient is fasting, supportive hydration by a stable intravenous (IV) line is commonly appropriate. Hydration is important for subsequent venous access and prevention of air embolization during the implant procedure. It can be frustrating and dangerous to try to gain venous access in a patient who is dehydrated after prolonged fasting without intravenous (IV) hydration. We generally request that the IV line be started on the side of the planned procedure to facilitate venography during attempts at venous access if this becomes a problem.

The management of CIED implantation in a patient taking anticoagulants is controversial, but recent studies indicate noninterruption in many situations is less risky than previously thought. Clearly, however, the patient receiving anticoagulants, especially heparin and/or platelet antagonists, is at risk for hematoma formation. Interruption of anticoagulation in some patients is hazardous (e.g., with artificial heart valve). Although historically many operators have chosen to “bridge” patients by using IV heparin or low-molecular-weight heparin subcutaneously while warfarin is withheld, there is an increasing trend toward noninterruption of oral anticoagulants. The recently published BRUISE CONTROL trial suggested that continuing warfarin through the CIED procedure process rather than bridging with heparin was associated with a lower incidence of wound hematomas (3.5% vs. 16.0%), including those that prolong hospitalization, require evacuation, and require interruption of anticoagulation with no increase in other morbidity. If bridging is used, the heparin is stopped on the day of surgery and the procedure is carried out. The heparin can be resumed several hours postoperatively, and warfarin can be started the same day or the next if there is no evidence of significant hematoma. When therapeutic levels are reached, the heparin is stopped, and the patient is managed only with oral warfarin therapy. With other oral anticoagulants currently in use (though not currently approved for use in patients with mechanical valves), shorter half-lives of the drugs allow for shorter preprocedural and postprocedural bridging. If we feel the need to bridge at our institution, our practice has been to stop the newer oral anticoagulants 2 to 3 days before the procedure and begin heparin bridging the next day, recognizing little data currently exist for this approach. One caveat with the newer oral anticoagulants is the current lack of “easy” reversal of anticoagulation compared with the use of vitamin K in patients receiving warfarin.

In 2008, the American College of Chest Physicians (ACCP) published evidence-based practice guidelines for the perioperative management of patients receiving antithrombotic therapy, including vitamin K antagonists (VKAs) and antiplatelet drugs. The ACCP recommends temporary cessation of VKAs and use of perioperative bridging anticoagulation with low-molecular-weight heparin and unfractionated heparin (UFH) for patients at moderate to high risk for thromboembolism and for those with a mechanical heart valve, AF, or venous thrombosis. There is no mention of uninterrupted VKA therapy. For the patient taking antiplatelet drugs who has bare metal or drug-eluting stents and requires surgery within 6 weeks of stent placement, uninterrupted antiplatelet therapy is recommended. In patients who require temporary interruption of antiplatelets, treatment is stopped 7 to 10 days before surgery. It is recommended that antiplatelet drugs be resumed approximately 24 hours postoperatively. Frequently with pacemaker and ICD procedures, however, antiplatelet therapy cannot be safely suspended.

Our long-term experience, now supported by the previously noted BRUISE CONTROL trial data, indicates that use of heparin seems to be more likely than administration of oral anticoagulants to result in hematomas. Our current practice in both new implants and generator replacement procedures is to continue oral anticoagulants, except in those patients in whom increased bleeding risks are anticipated (e.g., retropectoral procedures, procedures involving significant tunneling, in those involving lead extractions, and in some patients undergoing LV lead placement for CRT). Although there is growing support for this approach, overgeneralization would be inappropriate, and due consideration must be made in each patient on the basis of both current evidence and nuances in individual situations. Obviously, in some patients, especially those for whom oral anticoagulant therapy interruption briefly (especially the shorter-acting agent) is unlikely to result in thrombotic complications, such preprocedural interruption and postprocedural reinstitution may be completely appropriate.

With respect to other medications, generally patients are encouraged to continue all medications, on schedule, with sips of water. Reducing preoperative hypoglycemic agents by 50% has been our long-standing approach and seems to be successful. The administration of prophylactic antibiotics is increasingly standard practice, with recent data supporting. Although timing of periprocedural prophylactic antibiotics has been studied, there have been no clear specific conclusions, except that the 0- to 2-hour preprocedure window appears generally best. Typically, cephalosporins can be administered effectively in the hour before the procedure incision and vancomycin at the 1- to 2-hour preincision point. This has become a quality performance metric in the United States. Hospitals typically have standard preoperative regimens. We use preprocedural administration of a broad-spectrum cephalosporin (e.g., cefazolin) or vancomycin, which has the advantage of covering methicillin-resistant Staphylococcus. However, because cephalosporins may be more effective in preventing nonresistant staphylococcal infections, unless patients are penicillin-allergic or have existing methicillin-resistant infections, we continue to prefer cephalosporin use. Broad-spectrum “staph” coverage is important because of its preponderance in CIED infections. This is discussed in greater detail below. The patient should scrub the chest, neck, shoulders, and supraclavicular fossae with a povidone-iodine or chlorhexidine sponge the evening and the morning before the procedure. The surgical area usually is shaved in the procedure room. Informed consent should be attained consistent with facility policy and other regulatory considerations. Also, the patient should empty the bladder before coming to the procedure room.

Site Preparation and Draping

When effective patient support has been established, focus turns to the operative site. If not already accomplished, clipping of hair and skin cleansing should include the neck, supraventricular fossae, shoulders, femoral access sites, and chest. The operative site, hair clipped and cleansed, is now formally prepared and draped. Historically, and still commonly, a povidone-iodine scrub may be followed by alcohol, then povidone-iodine solution, with skin drying before applying the final povidone-iodine solution. Alternately, povidone-iodine gel is spread liberally over the operative site. Within 30 seconds, an optimal bactericidal effect is achieved. With this approach, scrubbing of the area is not required. For patients allergic to povidone-iodine, a chlorhexidine or hexachlorophene scrub can be used.

Many traditional scrubs have been replaced by povidone-iodine or a chlorhexidine and alcohol combination. These preoperative skin preparations have the benefit of a single, rapid application. DuraPrep is iodine povacrylex and isopropyl alcohol; ChloraPrep is 2% chlorhexidine and 70% isopropyl alcohol. Both offer rapid-acting, broad-spectrum protection. Because alcohol-based antiseptic solutions can act as fuel for surgical fires, the skin preparation must be allowed to dry, strictly observing recommended drying times. In addition, it is important to remove the fuel. Surgical fires in the operating room can result in patient burns and even death.

The draping process is a matter of personal preference. Use of a sterile, see-through plastic adhesive drape (with or without impregnation with an iodoform solution) over the entire operative area has become standard. After some form of sterile barrier is established, the operative site is draped with one or more large, sterile sheets. Care must be taken to avoid covering the face of the patient too closely, and this can be derived in a variety of ways. Many facilities have prefabricated tools to accomplish this, though a simple, cost-effective solution consists of a length of common house wire (8/3-gauge Romex) shaped into an arc over the patient's neck. The ends of the wire are bent at right angles to the arc and tucked under the x-ray table padding at the level of the patient's shoulders ( Fig. 26-3 ). The weight of the patient's shoulders supports the wire arc. The wire positioned under the shoulder is checked with fluoroscopy to avoid interference with the radiographic field of view.

From the moment the catheterization laboratory or special studies room is cleaned, it must be treated as a surgical suite. All personnel should wear surgical clothing, hats, and masks consistent with operating room standards.

Anesthesia, Sedation, and Pain Relief

Most CIED procedures can be performed with local anesthesia and some form of sedation and pain reliever. Local anesthesia alone is frequently inadequate for optimal patient comfort, clearly so for ICD implants in which defibrillation testing is performed. Its effect does not completely prevent the discomfort associated with creation of the CIED pocket. Therefore the additional combination of a narcotic and sedative is recommended; use of sedation alone is frequently inadequate. The challenge to the physician in charge is to achieve patient comfort without risking oversedation or respiratory depression. If an anesthesiologist or nurse anesthetist is part of the implantation team, patient comfort is usually achieved easily and safely. In this situation, if respiratory depression occurs, assisted ventilation is easily accomplished. When the implanting physician manages the sedation and narcotics, the patient must be carefully monitored by the circulating nurse. The medications should be administered slowly.

The selection and dose of local anesthetic are also important considerations. A local agent in therapeutic concentration that provides rapid onset of action and sustained duration is desirable. Local agents can be used in combination to achieve the desired effect, such as lidocaine for its rapid onset and bupivacaine for its sustained action. Also, the upper limit of total local anesthetic dose should not be exceeded. Toxic blood levels of local anesthetics can result in profound neurologic abnormalities, including obtundation and seizures. Table 26-1 lists the pharmacologic properties of common local anesthetic agents.

The selection of sedative and narcotic depends on personal preference. We use midazolam and fentanyl. The operator should become familiar with one or more sedative agents, as well as an analgesic, preferably a narcotic. Many newer agents are available. The selection of a benzodiazepine in combination with a semisynthetic narcotic can achieve ideal sedation, amnesia, and analgesia. Occasionally, diphenhydramine can be adjunctive. A cooperative, relaxed, and pain-free patient is fundamental to the success of the procedure and the avoidance of complications. Pentothal and nitrous oxide have been used to effect brief periods of complete sedation at times of anticipated maximum discomfort, but the use of these drugs requires the expertise of an anesthetist because temporary respiratory support is frequently needed.

The U.S. Joint Commission on Accreditation of Healthcare Organizations mandates that institutions establish a policy and protocol for patients receiving IV sedation, including CIED procedures. In essence, the protocol requires formal patient assessment before sedation. Resuscitation equipment must be present at all times in the sedation and recovery areas, and patients undergoing IV sedation must be monitored with pulse oximetry, continuous ECG rhythm monitoring, and automatic blood pressure recordings. Monitoring of the patient should continue for at least 30 minutes after the last IV sedative dose and for at least 90 minutes after intramuscular (IM) sedative administration. There are also strict discharge criteria. Table 26-2 lists common intravenous sedation drug protocols. A North American Society of Pacing and Electrophysiology (NASPE; now the Heart Rhythm Society) expert consensus developed recommendations and specified minimum training requirements on the use of IV sedation/analgesia by nonanesthesia personnel in patients undergoing arrhythmia-specific diagnostic, therapeutic, and surgical procedures.

Antibiotic Prophylaxis and Wound Irrigation

The use of prophylactic antibiotics (previously discussed briefly) to reduce the incidence of postoperative wound infection in a CIED procedure is currently standard practice. Importantly, antibiotics are not a substitute for good infection control practices, an adequate surgical environment, and good surgical technique. The use of antibiotics in a pacemaker procedure follows the principle of prophylaxis, in which the risk for infection is low but the morbidity is high. The selection of antibiotics is based on site-specific flora for wound infection and the spectrum, kinetics, and toxicity of the antimicrobial agent. The risk factors for infection have been well-defined. The National Research Council for Wound Classification places the risk for infection from an elective procedure with primary closure at less than 2%. One important consideration is the higher risk for infection in procedures lasting longer than 2 hours.

Now more formally studied (and previously discussed), the use of prophylactic antibiotics in the low-risk, high-morbidity group, such as patients receiving pacemakers and defibrillators, appears justified. A meta-analysis of antibiotic prophylaxis showed a significant reduction in the incidence of infection. The spectrum of the antibiotic prophylaxis generally needs to cover only the Gram-positive skin flora, primarily Staphylococcus epidermidis and S. aureus. In the case of CIEDs, the cephalosporins appear ideal (e.g., 2 g of cefazolin IV within 1 hour preincision). Because of increasing prevalence of methicillin-resistant S. aureus or S. epidermidis, vancomycin should be considered (e.g., 1 g IV slowly, at least 1 hour preincision). Postoperative doses are left to clinical judgment. Generally, 1 g of either drug may be given intravenously (IV) up to 8 hours postoperatively. Occasionally, the postoperative doses of cephalosporin are given orally for several days. A large, prospective, randomized double-blind, placebo-controlled trial (RCT) recently validated the efficacy of antibiotic prophylaxis before implantation of pacemakers and defibrillators. The trial planned to enroll 1000 patients, but enrollment was terminated at 649 patients by the safety committee. The study demonstrated a significant reduction in infectious complications with antibiotic prophylaxis with 1 g of cefazolin administered before the procedure. Pretreatment with cefazolin reduced the incidence of postprocedural infection (0.64% cefazolin vs. 3.28% placebo; P = .016).

The utility of prophylactic antibiotics has now been well established. Studies have shown that a single preoperative dose of antibiotics is as effective as a 5-day course of postoperative therapy. The prophylaxis should target the anticipated organisms. With complicated or contaminated procedures, additional postoperative coverage is indicated. During prolonged procedures, antibiotics should be readministered every 3 hours. Prophylactic antibiotics generally should be administered within 1 hour before incision and need not be given more than 24 hours after the procedure.

An additional strategy in the prevention of infection is topical antibiotic prophylaxis or antibiotic wound irrigation. Controlled trials evaluating the benefit of antibiotic irrigations are lacking. The concept of irrigation is to provide a high concentration of antibiotic at the site of potential infection at the time of contamination. The technique has proved most efficient in the absence of established infection and uses nonabsorbable antibiotics. Historically, aminoglycosides and bacitracin combinations have been used, but regimens vary in number, type, concentration, and duration of antibiotic use. Systemic toxicity with antibiotic irrigation is a concern, though, is rare. The superiority of irrigation over systemic antibiotic administration has never been proved, and even if used it is not a substitute for prophylactic preprocedural systemic antibiotic administration. Given the potential toxicity, caution in its use is recommended. Table 26-3 lists common antibiotic irrigation protocols.

Recently, a novel nonresorbable antibacterial pouch (AIGISx, TyRxPharma) was developed to inhibit biofilms of S. aureus on CIEDs. It consists of a controlled-release polypropylene envelope impregnated with the antibiotics rifampin and minocycline. Its purpose is to reduce the incidence of pocket infections associated with pulse generator changes and other CIED procedures in the patient at high risk for infection. It is not recommended for all patients. An in vitro study demonstrated that the envelope significantly reduced the ability of S. aureus to form biofilms on mock CIEDs. Such patients at high risk for infection include the immunocompromised patient with renal failure, patients receiving oral anticoagulants, and those undergoing CIED replacement/revision procedures. Bloom et al reported on use of the antibacterial pouch in 624 consecutive CIED procedures of high-risk patients. Device implantation was successful in 621 procedures, with only three major infections. There were no deaths related to the pouch. Use of the pouch was associated with high CIED implantation success with low infection rate in a population at high risk for CIED infection. No data show that the antibiotic pouch reduces the rate of clinical infection in any population. Resorbable antibiotic pouches are currently available and being formally investigated in randomized clinical trials for their impact on CIED infection in the WRAP-IT trial and the preoperative, intraoperative, and postoperative antibiotic regimens in a randomized cohort study called PADIT. These trials should both determine the value of these interventions (there is substantial additional expense, currently, for these new pouches) and also determine the actual rates of infections during CIED procedures.

Anatomic Approaches for Implantation

There are two basic anatomic approaches to the implantation of a CIED that are described in this chapter. There are two other approaches—subcutaneous ICDs and leadless pacemakers—that are discussed in other chapters. Historically, the first is the epicardial approach and the second, the transvenous approach. The epicardial approach calls for direct application of electrodes on the heart. This requires general anesthesia and surgical access to the epicardial surface of the heart. The transvenous approach is usually performed with local anesthesia and IV sedation. Each approach can be accomplished by several unique techniques. Currently, more than 95% of all pacemaker implants and virtually all nonsubcutaneous ICD implants are performed transvenously. The epicardial approach is generally reserved for patients who cannot undergo safe or effective pacemaker implantation by the transvenous route. The major epicardial techniques involve either applying the electrode(s) directly to a completely exposed heart or performing a limited thoracotomy through a subxiphoid incision ( Fig. 26-4 ). A third technique places the leads by mediastinoscopy. There is even a fourth technique, which combines epicardial and endocardial lead placement. These techniques overlap the techniques used for surgical epicardial LV leads for CRT described in Chapter 31 . The several techniques used for the transvenous approach involve a venous surgical cutdown, percutaneous venous access, or a combination of both ( Box 26-2 ). The pros and cons of the various approaches and techniques are reviewed here. Whereas ICDs were initially implanted almost exclusively by epicardial techniques, the development of intravascular defibrillation coils has favored conversion to transvenous implant techniques.

A thorough knowledge of the anatomic structures of the neck, upper extremities, and thorax is essential for CIED implantation ( Fig. 26-5 , Video 26-1 ). The precise location and orientation of the internal jugular, innominate, subclavian, and cephalic veins are important for safe venous access. Their anatomic relationships to other structures are crucial to avoiding complications.

The venous anatomy of interest, from a cardiac pacing point of view, starts peripherally with the axillary vein. This large venous structure represents the continuation of the basilic vein and starts at the lower border of the teres major tendon and latissimus dorsi muscle. The axillary vein terminates immediately beneath the clavicle at the outer border of the first rib, where it becomes the subclavian vein. The axillary vein is covered anteriorly by the pectoralis minor and pectoralis major muscles and the costocoracoid membrane. The axillary vein is anterior and medial to the axillary artery, which it partially overlaps. At the level of the coracoid process, the axillary vein is covered only by the clavicular head of the pectoralis major muscle ( Fig. 26-6 ). At this juncture, the axillary vein receives the more superficial cephalic vein.

The cephalic vein terminates in the deeper axillary vein at the level of the coracoid process beneath the pectoralis major muscle. The cephalic vein, sometimes used for CIED venous access, is classified as a superficial vein of the upper extremity. This vein, which actually commences near the antecubital fossa, travels along the outer border of the biceps muscle and enters the deltopectoral groove, an anatomic structure formed by the deltoid muscle and clavicular head of the pectoralis major. The cephalic vein traverses the deltopectoral groove and superiorly pierces the costocoracoid membrane, crossing the axillary artery and terminating in the axillary vein just below the clavicle at the level of the coracoid process.

The subclavian vein is a continuation of the axillary vein. The subclavian vein extends from the outer border of the first rib to the inner end of the clavicle, where it joins with the internal jugular vein to form the brachiocephalic or innominate vein. The subclavian vein is just inferior to the clavicle and subclavius muscle. The subclavian artery is located posterior and superior to the vein. These two structures are separated internally by the scalenus anticus muscle and phrenic nerve. Inferiorly, the subclavian vein is associated with a depression in the first rib and on the pleura. The brachiocephalic or innominate veins are two large, venous trunks located on each side of the base of the neck. The right innominate vein is relatively short. It starts at the inner end of the clavicle and passes vertically downward to join with the left innominate vein just below the cartilage of the first rib to form the superior vena cava (SVC). The left innominate vein is larger and longer than the right, passing from left to right for approximately 2.5 inches (6 cm), where it joins with the right innominate vein to form the SVC. The left innominate vein is in the anterior and superior mediastinum.

The internal and external jugular veins have also been used for CIED venous access. The external jugular vein is a superficial vein of the neck that receives blood from the exterior cranium and face. This vein starts in the substance of the parotid gland, at the angle of the jaw, and runs perpendicular down the neck to the middle of the clavicle. In this course, the external jugular crosses the sternocleidomastoid muscle and runs parallel to its posterior border. At the attachment of the sternocleidomastoid to the clavicle, the external jugular vein perforates the deep fascia and terminates in the subclavian vein just anterior to the scalenus anticus muscle. The external jugular is separated from the sternocleidomastoid muscle by a layer of deep cervical fascia. Superficially, it is covered by the platysma muscle, superficial fascia, and skin. The external jugular vein can vary in size and may even be duplicated. Because of its superficial orientation, the external jugular vein is less frequently used for CIED venous access ( Fig. 26-7 ).

The internal jugular vein is an unusual site for CIED venous access. Because of its larger size and deeper and more protected orientation, however, the internal jugular vein is used more frequently than the external jugular vein. The internal jugular vein starts just external to the jugular foramen at the base of the skull. It drains blood from the interior of the cranium, as well as superficial parts of the head and neck. This vein is oriented vertically as it runs down the side of the neck. Superiorly, the internal jugular is lateral to the internal carotid and inferolateral to the common carotid. At the base of the neck, the internal jugular vein joins the subclavian vein to form the innominate vein. The internal jugular vein is large and lies in the cervical triangle, defined by the (1) lateral border of the omohyoid muscle, (2) inferior border of the digastric muscle, and (3) medial border of the sternocleidomastoid muscle. The superficial cervical fascia and platysma muscle cover the internal jugular vein, which is easily identified just lateral to the easily palpable external carotid artery.

From a venous access perspective, the location of the subclavian vein may vary from a normal lateral course to an extremely anterior or posterior orientation in elderly patients. Byrd has described the subclavian venous anatomy of two distinct deformities, both of which make venous access more difficult and hazardous. The first deformity involves a posteriorly displaced clavicle ( Fig. 26-8 ). This is usually seen in patients with chronic lung disease and anteroposterior chest enlargement. Such patients can be identified by the presence of a horizontal deltopectoral groove and the posteriorly displaced clavicle. The second deformity is an anteriorly displaced clavicle ( Fig. 26-9 ), which is found occasionally, especially in elderly women. In this situation, the clavicle is anteriorly bowed or actually displaced anteriorly. It is important that the implanting physician recognize such variations to avoid complications such as pneumothorax and hemopneumothorax when using the percutaneous approach.

It is assumed that the implanting physician is also completely familiar with the anatomy of the heart and great vessels. However, their spatial orientation is at times confusing, particularly with respect to the right atrium (RA) and right ventricle (RV). In the frontal plane, the border of the right side of the heart is formed by the RA. The border of the left side of the heart is composed of the left ventricle. Importantly, the RV is located anteriorly ( Fig. 26-10 ) and is triangular. The apex of the RV is the generally accepted initial target for ventricular lead placement, although its location can vary. Its normal location, distinctly to the left of midline, depends on the rotation of the heart, which is affected by various pathologic and anatomic conditions. At times, the apex may be located directly anterior to or even to the right of midline. A lack of appreciation of these variations can lead to considerable difficulty in electrode placement.

The choice of site for CIED implantation is also occasionally important anatomically. This decision is typically made most appropriately on the basis of the patient's dominant hand, occupation, recreational activities, and medical conditions. The decision should not be made according to the dominant hand of the implanting physician. However, some fundamental differences exist between the anatomy of the right and left sides, which can be frustrating when passing a CIED lead. It seems to be easier for many right-handed implanters to work on the right side of the patient, and vice versa, but from a surgical point of view, catheter manipulation from the right can be a frustrating experience. When entering the central venous circulation from the left upper limb, the lead tracks along a smooth arc to the RV. There are generally no sharp angles or bends ( Fig. 26-11A ). Conversely, when the approach is from the right, the lead is forced to negotiate a sharp angle or bend at the junction of the right subclavian and internal jugular veins, where the innominate vein is formed ( Fig. 26-11B ). This acute angulation can make the manipulation of the lead more difficult when a stylet is fully inserted. Another anatomic pitfall occurs when there is a persistent left SVC, making passage to the heart from the left more difficult and, if there is no right SVC, making passage from the right impossible. These situations are considered later, in the discussion of ventricular electrode placement.

Cephalic Venous Access

The right or left cephalic vein is the most common vascular entry site for insertion of CIED leads by the cutdown technique. The cephalic vein is located in the deltopectoral groove ( Fig. 26-12 ), which is formed by the reflections of the medial head of the deltoid and the lateral border of the greater pectoral muscles. The groove can be precisely located by palpating the coracoid process of the scapula. The dermis along the deltopectoral groove is infiltrated with local anesthetic, encompassing the anticipated length of the incision. A vertical incision is made adjacent to and at the level of the coracoid process. It is extended for about 2 to 5 cm. Care is taken to keep the scalpel blade perpendicular to the surface of the skin. One can create smooth skin edges by making an initial single stroke that carries through the dermis to each corner of the wound. The subcutaneous tissue is infiltrated with local anesthetic along the edges of the incision. The Weitlaner retractor is applied to the edges of the wound, and the subcutaneous tissue is placed under tension. The tension is released by light strokes of the scalpel (or electrosurgical tool) from corner to corner of the wound in the midline. As the subcutaneous tissue falls away, tension is restored by reapplication of the Weitlaner retractor. This process is continued down to the surface of the pectoral fascia. The fascia is left intact. At this level, the borders of the pectoral and deltoid muscles forming the deltopectoral groove are identified. A Metzenbaum scissors is used to dissect along the groove by separating the muscles' fibrous attachments. The Weitlaner retractor is reapplied more deeply to retract the muscle. Gradual release of the fascial tissue between the two muscle bodies will expose the cephalic vein.

At times, the cephalic vein is diminutive or atretic and unable to accommodate a CIED lead. In this case, the cephalic vein can be dissected centrally to the axillary vein, and this larger vein can be catheterized. Once the vein to be catheterized is localized, it is freed of all fibrous attachments. Ligatures are applied proximally and distally ( Fig. 26-13A ). The distal ligature is tied and held by a small clamp. The proximal ligature is not tied but is kept under tension with another clamp. An arbitrary entry site is chosen between the two ligatures. The anterior half of the vein at this site is grasped with a smooth forceps, and the vein is gently lifted. A small, horizontal venotomy is made with iris scissors ( Fig. 26-13B ) or a No. 11 scalpel blade. The vein is continuously supported by the forceps. The venotomy is held open by any of several means: a mosquito clamp, forceps, or vein pick. Gentle traction is applied on the distal ligature while tension is released on the proximal ligature. With the venotomy held widely open, the lead or leads are inserted and advanced into the central venous circulation ( Fig. 26-13C ).

Subclavian Venous Access

For many years, vascular access has been achieved for many purposes through the use of the Seldinger technique. This simple approach calls for the percutaneous puncture of the vessel with a relatively long, large-bore needle; passage of a wire through the needle into the vessel; removal of the needle; and passage of a catheter or sheath over the wire into the vessel with removal of the wire. An 18-gauge, thin-walled needle 5 cm in length is typically used, although smaller needles are available. These needles come prepackaged with most introducer sets, but an extra supply should be available. The historical problem limiting the use of this technique for CIED implantation was the inability to remove the sheath from the lead. The development of a peel-away sheath by Littleford solved this problem.

Use of the percutaneous approach requires a thorough knowledge of both normal and abnormal anatomy to avoid complications. For many years, the subclavian vein was generally the intended venous structure used for percutaneous venous access for CIEDs. Given the previously discussed anatomic variations, the subclavian vein puncture is typically made near the apex of the angle formed by the first rib and clavicle. This defines the “subclavian window” ( Fig. 26-14 ). At this puncture site (and after both skin infiltration with local anesthetic and a 1-cm incision at the site, which generally is 1-2 cm inferolateral to the point where the clavicle and first rib actually cross), the needle is aimed in a medial and cephalic direction. It is important to make the puncture with the patient in a “normal” anatomic position. The infraclavicular space or costoclavicular angle should not be artificially opened by maneuvers such as extending the arm or placing a towel roll between the scapulae. These maneuvers can open a normally closed or tight space and lead to undesirable puncture of the costoclavicular ligament or subclavius muscle, which in turn can result in lead entrapment and crush. With the patient in the normal anatomic position, access to the subclavian window is medial yet usually avoids the costoclavicular ligament. The more medial puncture and needle trajectory of this approach vastly improves the success rate and dramatically reduces the risks of pneumothorax and vascular injury compared with a more lateral approach. With this medial position, the vein is a much larger target and the apex of the lung is more lateral. This safer approach is a departure from the conventional subclavian venous puncture, which calls for introduction of the needle into the middle third of the clavicle.

There are legitimate concerns that this medial approach, although safer, results later in higher complication rates and failure rates due to conductor fracture and insulation damage. It is postulated that the extreme medial position results in a tight fit, subjecting the lead to compressive forces and causing binding between the first rib and the clavicle. Occasionally, this binding can even crush the lead, referred to as the subclavian crush phenomenon. This phenomenon is more common in larger, complex leads of the in-line bipolar, coaxial design. Fortunately, the incidence of this complication is low. Fyke first reported insulation failure of two leads placed side by side with use of the percutaneous approach through the subclavian vein, where there was a tight costoclavicular space. This issue has now been addressed thoroughly by two independent groups. Jacobs et al analyzed a series of failed leads for the mechanism of failure, using autopsy studies to correlate the anatomic relationship of lead position to compressive forces ( Fig. 26-15 ). These autopsy data demonstrated generation of significantly higher pressure when leads were inserted in the costoclavicular angle than with a more lateral puncture. The authors concluded that the tight costoclavicular angle should be avoided. Magney et al derived similar data from cadaveric studies and suggested that lead damage is caused by soft tissue entrapment by the subclavius muscle rather than by bony contact. This soft tissue entrapment causes a static load on the lead at that point, and repeated flexure around the point of entrapment may be responsible for the damage.

Concern about subclavian crush has also been communicated by CIED manufacturers in company literature. Reduction in lead diameter and modification of lead technology have not eliminated this problem. However, technique modification appears to be effective at reducing its occurrence. If a pacemaker lead feels tight in the costoclavicular space, it is likely more susceptible to being crushed and removing the lead from the vein in this situation and repuncturing the vein in a different location with reintroduction of the lead is advisable. Although this practice reduced the incidence of crush, more substantial modifications in technique, described later, seem to have essentially eliminated the crush problem.

Addressing this issue, along with other introducer-related or percutaneous complications, Byrd described a “safe introducer technique.” This technique consists of a “safety zone” associated with precise conditions ensuring a safe puncture. Byrd's safety zone is defined as a region of venous access between the first rib and the clavicle, extending laterally from the sternum in an arc ( Fig. 26-16A ). As a condition for puncture, the site of access must be adequate for ease of insertion to avoid friction and puncture of bone, cartilage, or tendon. With this technique, subclavian vein puncture should never be made outside the safety zone or in violation of the preceding conditions.

Byrd also has described a new technique for cannulating the axillary vein that may now be the dominant technique used (see below). As previously mentioned, the axillary vein is actually a continuation of the subclavian vein after it exits the superior mediastinum and crosses the first rib. The axillary vein is also frequently referred to as the extrathoracic portion of the subclavian vein ( Figs. 26-16B , and 26-17 ).

Axillary Venous Access

The axillary vein approach is actually not new. In 1987, on the basis of cadaveric studies that established reliable surface landmarks, Nichalls and Taylor and Yellowlees reported this approach as an alternative safe route of venous access for large, central lines. The axillary vein has a completely infraclavicular course ( Fig. 26-18 ). The needle path must always be anterior to the thoracic cavity, avoiding risks of pneumothorax and hemothorax (see Box 26-2 .)

The axillary vein starts medially at a point below the aspect of the clavicle where the space between the first rib and the clavicle becomes palpable. The vein extends laterally to a point about three fingerbreadths below the inferior aspect of the coracoid process. The skin is punctured along the medial border of the smaller pectoralis muscle at a point above the vein as it is defined by the surface landmarks. One punctures the axillary vein by passing the needle anterior to the first rib, maneuvering posteriorly and medially corresponding to the lateral to medial course of the axillary vein. The needle never passes between the first rib and the clavicle but stays lateral to this juncture. Some implanters have found it useful to abduct the arm 45 degrees when using this approach.

In the technique described by Byrd, the axillary vein puncture is performed as a modification of the standard subclavian vein procedure without repositioning of the patient ( Fig. 26-19 ; see also Fig. 26-16B ). The introducer needle is guided by fluoroscopy directly to the medial portion of the first rib. The needle is held perpendicular to, and touches, the first rib. The needle, held perpendicular to the rib, is “walked” laterally and posteriorly, touching the rib with each change of position. Once the vein is punctured, as indicated by aspiration of venous blood into the syringe, the guidewire and the introducer are inserted with use of standard technique. This approach essentially guarantees a successful and safe venipuncture without compromising the leads if the conditions for entering the safety zone are adhered to and if the first rib is touched to maintain orientation. The only complication not prevented by this approach is inadvertent puncture of the axillary artery.

Byrd has reported success in a series of 213 consecutive cases in which the extrathoracic portion of the subclavian vein (axillary vein) was successfully cannulated as a primary approach. Magney et al subsequently reported a new approach to percutaneous subclavian venipuncture to avoid lead fracture. This technique is very similar to Byrd's and uses extensive surface landmarks for venipuncture ( Fig. 26-20 ). It involves puncture of the extrathoracic portion of the subclavian vein. Magney et al define the location of the axillary vein as the intersection with a line drawn between the middle of the sternal angle and the tip of the coracoid process. This is generally near the lateral border of the first rib.

Belott described blind axillary venous access using a modification of the Byrd and Magney recommendations. In this technique, the deltopectoral groove and coracoid process are primary landmarks and are palpated, and the curvature of the chest wall is noted ( Fig. 26-21 ). An incision is made at the level of the coracoid process. It is carried medially for about 6.25 cm (2.5 inches) and is perpendicular to the deltopectoral groove ( Fig. 26-22 ). The incision is carried to the surface of the pectoralis major muscle. The deltopectoral groove is visualized on the surface of the muscle. The needle is inserted at an angle 45 degrees to the surface of the pectoralis muscle and parallel to the deltopectoral groove and 1-2 cm medial ( Fig. 26-23 ). Currently, both authors of this chapter use Byrd's first rib/fluoroscopy approach because it essentially eliminates risk of pneumothorax. Critical to its success is the ability to identify the first or second rib on a radiograph. In addition, if the first rib is poorly visualized or set too far under the clavicle, the second rib can be used in a similar fashion.

To access the axillary vein using the first rib, fluoroscopy is used to image the medial shoulder area. The first rib is then identified, usually the most superior U-shaped rib ( Fig. 26-24A ). The ribs seen traversing medial to lateral in an inferior direction are posterior. Identifying the first rib fluoroscopically is critical. If the operator misinterprets a posterior rib as the first rib, a percutaneous stick will result in a pneumothorax or access of undesired cardiopulmonary structures. The first step in accessing the axillary vein using the first rib is to place the 18-G percutaneous needle and syringe on top of the pectoralis major muscle in the superior aspect of the incision or previously formed CIED pocket. Using fluoroscopy, the needle tip is placed in the middle of the first rib ( Fig. 26-24B ). The angle of the syringe and needle is gradually increased as the needle is advanced through the pectoralis major muscle. The forward motion of the percutaneous needle and syringe should allow the tip of the needle to be maintained fluoroscopically over the body of the first rib. To maintain first rib orientation, a rather steep angle is generally required. The needle advancement is continued until the first rib is struck. In essence, this maneuver is attempting to pin the axillary vein to the first rib ( Fig. 26-25 ). Once the first rib is touched, the needle and syringe are slowly withdrawn under suction until the vein is entered, as indicated by a flash of blood in the syringe. If the first pass is unsuccessful, the needle and syringe are moved medially or laterally, and the maneuver is repeated until successful. Once the vein is entered, the guidewire is passed and the sheath applied per standard technique. If the needle is advanced toward the first rib through tissue or muscle without the needle tip initially visualized fluoroscopically directly over the first rib, this shallow angle may result in the needle passing through an intercostal space. This can result in a pneumothorax. It is recommended that a figure-eight or purse string stitch be applied around the needle puncture in the muscle for hemostasis. Multiple punctures or the retained-guidewire technique can be used for multiple lead placements ( Video 26-2 ).

Occasionally, in a thin patient, the axillary artery can be easily palpated. This makes the axillary vein stick easy because the percutaneous puncture can be made just medial and inferior to the palpable axillary pulse. Because it cannot always be palpated, the axillary pulse is not a reliable landmark. The axillary artery and brachial plexus are usually much deeper and more posterior structures. With a thorough knowledge of the regional anatomy, the physician can safely use the axillary vein as a primary site for venous access.

Access to the axillary vein may also be achieved by direct cutdown. With Metzenbaum scissors, fibers of the pectoralis major muscle are separated adjacent to the deltopectoral groove at the level of the coracoid process. This is just above the level of the superior border of the pectoralis minor. If the pectoralis major is split in this area and the fibers are gently teased apart in an axis parallel to the muscle bundle, the axillary vein can be found directly beneath the pectoralis major muscle. A purse-string stitch is applied to the axillary vein, which can then be cannulated by a direct puncture or cutdown technique. The purse-string stitch will serve for hemostasis and ultimately assists in anchoring the electrodes after positioning.

A variety of other techniques can facilitate access to the axillary vein. Varnagy et al describe a technique for isolating the cephalic and axillary veins by introduction of a radiopaque J -tipped polytetrafluoroethylene guidewire through a vein in the antecubital fossa under fluoroscopic control ( Fig. 26-26 ). The metal guidewire is then palpated in the deltopectoral groove or identified with fluoroscopy. This guides the subsequent cutdown or puncture of the vessel with fluoroscopy. A cutdown can be performed on a vein, and the intravascular guidewire pulled out of the venotomy to allow the application of an introducer. If a percutaneous approach is used, the puncture can always be extrathoracic, using fluoroscopy to guide the needle to the guidewire.

Contrast venography, described subsequently, can also be used for axillary venous access. The venous anatomy can be observed with contrast fluoroscopy in the pectoral area and, if possible, recorded for repeat viewing. The needle trajectory and venipuncture are guided by the contrast material in the axillary vein. Laboratories with sophisticated imaging capabilities can create an image “mask” (see below). Spencer et al reported the use of contrast material for localizing the axillary vein in 22 consecutive patients. Similarly, Ramza et al demonstrated the safety and efficacy of using the axillary vein for placement of pacemaker and defibrillator leads when guided with contrast venography. They successfully accomplished lead placement in 49 of 50 patients using this technique.

Access to the axillary vein can also be guided by Doppler flow detection and ultrasound techniques. Fyke described use of an extrathoracic introducer insertion technique in 59 consecutive patients (total of 100 leads) with a simple Doppler flow detector. A sterile Doppler flow detector is moved along the clavicle. Once the vein is defined, the location and angulation of probe are noted, and the venipuncture is carried out ( Fig. 26-27 ). Care is taken to avoid directing the Doppler beam beneath the clavicle. Gayle et al developed an ultrasound technique that directly visualizes the needle puncture of the axillary vein. A portable ultrasound device with sterile sleeve and needle holder are used. The ultrasound head is placed over the skin surface in the vicinity of the axillary vein. Once the vein is visualized, the puncture technique can be used ( Video 26-3 ). This technique has been used with considerable success for both pacing and defibrillator leads. There have been no reports of pneumothoraces. This technique can be carried out transcutaneously or through the incision on the surface of the pectoralis muscle ( Figs. 26-28 and 26-29 ).

In summary, the axillary vein has become a common, and probably the dominant, venous access site for CIED implantations, because of concerns about subclavian crush, pneumothorax, and the requirement for insertion of multiple leads. A number of reliable techniques are available for axillary venous access (see Box 26-2 ). Given the recent interest in the axillary vein, it is recommended that the implanting physician become thoroughly familiar with the relevant anatomy. Although extensive use of the axillary vein for CIED lead insertion seems to be quite successful and complication-free, there is also some concern that lateral access of the axillary vein can result in acute bends in the lead. This can result in increased lead stress and potential for fracture or lead damage because in its lateral aspect the axillary vein is a much deeper structure as it transitions to the subclavian vein. It is this deeper venous access that results in acute lead angulation. This is particularly true for those who use the second rib as a landmark for axillary venous access.

For more conventional subclavian vein approaches, Lamas et al recommended fluoroscopic observation of the needle trajectory to achieve a successful and safe subclavian vein puncture. They initially identify the clavicle on the side of the puncture, noting its course and landmarks. The skin is entered about 2 cm inferior to the junction of the medial and lateral halves of the clavicle, aiming with fluoroscopic guidance for the caudal half of the clavicular head.

The various techniques of transvenous lead placement—the subclavian window, the safety zone, the axillary vein puncture, or fluoroscopic guidance—have some common features. Needle orientation is always medial and cephalad, almost tangential to the chest wall. All needle-probing should be done in a forward motion. Lateral needle-probing should be avoided because it could lacerate important structures. Anatomic landmarks are defined, and the puncture is made, with rare exception, with the patient in the anatomic position. The costoclavicular angle is not artificially opened by maneuvers. Although the essence of the nonaxillary approaches is medial placement to avoid the lung, the undesirable puncture of the costoclavicular ligament should be avoided. In the obese patient, the tendency is to orient the needle more perpendicular to the chest wall in an attempt to pass between the clavicle and first rib. This perpendicular angle is to be avoided with the medial approaches because it is associated with a higher incidence of pneumothorax. In this circumstance, a more inferior skin puncture is recommended, allowing the needle to slip between the first rib and clavicle. The needle is therefore kept almost tangential to the chest wall, avoiding the lung. Some implanters bend the needle in an attempt to slip under the clavicle. We do not recommend this maneuver, because it is associated with a higher incidence of pneumothorax and vascular trauma. We instead recommend that, in the morbidly obese patient, the subclavian puncture be carried out after direct visualization of the pectoral muscle. This can be done only by making an initial skin incision and carrying it down to the pectoral muscle. Once the anatomic landmarks are defined, the needle is slipped between the first rib and the clavicle with a trajectory that is nearly tangential to the chest wall and directed cephalad and medial.

This last recommendation raises the question whether the skin incision should routinely be made first, with percutaneous venous access carried out through the incision, or whether an initial percutaneous venous puncture should be performed, followed by the incision. It is arguably better not to commit to an initial pocket-length incision and subsequent venipuncture. Doing so avoids the embarrassment of having to explain bilateral incisions if venous access could not be achieved through the initial incision and the surgeon is forced to move to the patient's other side. As an alternative, a 1-cm-long stab wound can be made initially, through which the venipunctures can be accomplished. This approach allows easy incorporation of the puncture sites (especially if multiple separate punctures are used for a multilead system) into a single incision that is extended after successful venipuncture. Having described these more conservative approaches, making a full-length incision or even making the CIED pocket before venous access is a reasonable approach, especially for experienced implanters with experience in a variety of venous access techniques.

As with axillary vein access techniques, the subclavian puncture can be facilitated by the use of contrast venography. It is helpful in patients with potentially difficult venous access. Venography should be considered before any puncture in which venous patency is in doubt or abnormal anatomy is suspected. As described by Higano et al, a venous line is established in the arm on the side of planned pacemaker venous access. The line should be reliable and 20 gauge or larger. One must ensure that the patient does not have an allergy to radiographic contrast material. The contrast injection is performed by a nonsterile assistant. From 10 to 50 mL of contrast material is injected rapidly into the IV line in the forearm, followed by a saline bolus flush. The contrast medium moves slowly in the peripheral venous system and can be moved along by massage of the arm through or under the sterile drapes. The venous anatomy is observed with fluoroscopy in the pectoral area and, if possible, recorded for repeated viewing ( Fig. 26-30 ). The needle trajectory and venipuncture are guided by the contrast material in the subclavian vein. With more sophisticated but increasingly common radiologic systems, a mask or map can be made for guidance after the contrast medium has dissipated. The process can be repeated as necessary. An alternative when venous access is not possible on the side of planned lead implantation is to use of a catheter advanced from the femoral vein to the axillary vein over a wire to image the vessel with contrast. When a contrast allergy makes its use inadvisable if the patient is not adequately prepped for the allergy, positioning a guidewire from the femoral vein to the axillary vein provides a target for vascular access.

The actual venous puncture is carried out with a syringe attached to the 18-gauge needle. A common practice is to fill the syringe partially with saline. The theory behind this practice is that if a pneumothorax occurs, it will be detected by air bubbles aspirated through the saline. In addition, the saline can be used to flush out tissue plugs that may obstruct the needle and prevent aspiration. We avoid this practice because we believe that one does not need air bubbles to detect an inadvertent pneumothorax. More importantly, a syringe even partially filled with saline makes it more difficult to differentiate between arterial and venous blood, because when blood (arterial or venous) mixes with the saline it takes on the color of oxygenated blood. If saline is not used, the vascular structure that has been entered is more readily apparent.

When proceeding with a venipuncture, one should hold the syringe in the palm of the hand with the dorsal aspect of the hand resting on the patient. This gives support and control as the needle is advanced. With the needle held this way, tactile sensation is enhanced, and one can frequently feel the needle enter the vein. Once the vessel is entered, the guidewire is inserted and the tip advanced to a position in the vicinity of the right atrium ( Fig. 26-31 ). We prefer to use J -shaped or curved-tip guidewires for safety reasons. If resistance is encountered, the wire is withdrawn slightly and advanced again. If the resistance persists, the wire position is checked with fluoroscopy. If the wire just outside the tip of the needle appears coiled, it is probably extravascular. In this case, the wire and needle are removed and a new venous puncture is carried out. Rarely, the vein may not be able to be reentered. This may be due to collapse of the vein by a resultant hematoma caused by a small tear in the vein from the misdirected guidewire. In this case, one should probably proceed to an alternative approach or site of venous access to avoid an unnecessary waste of time and a higher risk of pneumothorax with multiple subsequent unsuccessful percutaneous punctures.

Occasionally, the guidewire tracks up the internal jugular vein. Changing the angle of the needle slightly to a more medial and inferior direction while the guidewire is still in the internal jugular vein, withdrawing the guidewire back into the needle, and then advancing again usually results in passage of the guidewire through the innominate vein and SVC into the right atrium. This maneuver may have to be repeated several times with varying needle angulations. Care must be exercised to avoid tearing the vein. The application of a 4-6-French (Fr) dilator and/or angled catheter can sometimes help steer the guidewire in the right direction. In rare cases, the guidewire and needle must be removed and a new puncture site selected. The key point is that once venous access has been achieved, every effort is made to retain it.

When air is withdrawn through the needle during attempted venipunctures, suggesting lung puncture and raising the possibility of a pneumothorax, our practice is to withdraw the needle; wait to ensure a rapid-onset, large, symptomatic pneumothorax is not occurring; and then proceed with a different needle trajectory and further attempts at venipuncture. In our experience, most lung punctures occurring with forward (not lateral!) needle motion do not result in a clinically apparent (on chest radiography) pneumothorax. If a pneumothorax does develop, it may do so in this setting over a matter of hours and may not even be apparent radiographically at the end of the procedure. If a lung puncture has occurred, obtaining another upright chest radiograph 6 hours after completion of the procedure is advisable. If a pneumothorax has developed, a chest tube or catheter evacuation procedure may be necessary, although frequently a small to moderate pneumothorax that is not expanding can be managed conservatively without evacuation.

Similarly, if an arterial puncture occurs inadvertently, our approach consists of removal of the needle and compression at the puncture site for about 5 minutes, followed by repeated venipuncture attempts with a different needle path. It is rare for such arterial punctures to result in a hemothorax, provided that no tearing of the artery has occurred; avoidance of lateral needle motion is crucial here as well. Follow-up chest radiographs are recommended 6 to 18 hours after the procedure, and postoperative hemoglobin and hematocrit measurements are suggested. The most important problem to avoid if the artery has been punctured is nonrecognition and placement of a sheath into the artery. If there is any doubt about whether the artery has been punctured, a blood sample withdrawn through the needle and subjected to oximetric analysis should clarify the situation.

As previously mentioned, once the wire is successfully positioned in the vein, some implanters place a purse-string suture in the tissue around the point of entry of the wire into the tissue. Alternatively, a figure-eight stitch can be applied ( Fig. 26-32 ). This step can be helpful later for hemostasis. Such sutures require that an incision be made beginning at the needle and extended posteriorly to the depth of the pectoral fascia.

Once the guidewires are in the subclavian vein, it is usually simple to advance the appropriate-sized dilator and peel-away sheath over the wire into the venous circulation. Occasionally, there is substantial resistance to dilator sheath advancement, and repetitive dilation with progressively larger dilators is necessary. Alternatively, a 15- to 20-degree bending of the tip of the full-sized introducer may facilitate advancement over the wire. If difficulty with advancement occurs, we generally remove the sheath from the dilator and use only the dilator initially to dilate the track into the vein. This protects the rather delicate sheaths from damage. After successful advance of the dilator alone over the wire, the dilator can be withdrawn, the sheath added, and both then advanced over the wire. We have found that gentle back-pressure on the guidewire while the dilator sheath is advanced also facilitates advancement in difficult or tortuous vessels.

After the sheath has been successfully passed over the guidewire to the vicinity of the SVC, the dilator and guidewire can be removed, and the lead advanced through the sheath. Problems can be encountered when the lead is passed through the sheath. Occasionally, in the process of introduction, the sheath buckles at a point in the venous system where there is a bend ( Fig. 26-33 ). This usually occurs after the removal of the dilator. It can also occur if the sheath is advanced against the lateral wall of the SVC; if a buckle occurs, the lead will not pass this point. Forcing the lead can result in damage to the tip electrode and insulation. This kink can usually be observed on fluoroscopy. There are several solutions to this problem. If the guidewire and dilator have both been removed, both can be replaced down the sheath. The dilator with wire inside is now functioning not only as a way to stiffen the sheath but also as a tip occluder, and both can be passed back down the sheath. The tapered tip of the dilator will straighten the buckle. One can change the position of the buckle point by slightly advancing or retracting the sheath. The dilator is removed, but the guidewire can be retained. It is hoped that the retained guidewire will act as a stent, preventing the buckle from recurring and thus allowing the lead to pass completely down the sheath. Another option when buckling of sheath occurs is to advance the lead to within a couple of centimeters of the buckle and then slowly withdraw the sheath, holding the lead position stationary. The sheath, including the buckle point, may occasionally be easily withdrawn over the tip of the lead, and the lead can be cautiously advanced. In this situation, it is sometimes necessary to withdraw the stylet from the tip of the lead to allow easy advancement of the lead beyond the buckle point. Inexperienced implanters should be cautious with this latter technique, however, because it may damage the distal electrodes.

If these maneuvers fail, the guidewire and dilator are reinserted and the sheath and dilator are removed, leaving the guidewire in the vein. Tissue compression or traction is applied to the purse-string suture for hemostasis. The sheath is inspected for buckling. At this point, the implanter should consider advancing a sheath of the next larger size over the retained guidewire, because application of the same-sized sheath usually results in recurrence of the same problem. The important point is that despite this frustrating experience, one must be reluctant to relinquish the vein once it has been catheterized. Generally, the larger sheath is less likely to buckle, especially if the guidewire is retained to act as a stent. With successful lead introduction, the sheath is pulled back out of the circulation, and skin compression or traction is applied to the purse-string or figure-eight suture for hemostasis.

The risk of air embolization is substantial with the introducer approach. Use of the Trendelenburg position or elevation of the patient's legs on a wedge or pillows should be considered. Newer peel-away sheaths with hemostatic valves substantially reduce but do not eliminate this risk, and their use has become standard. Adequate hydration serves both to reduce the risk of air embolism and to make venous punctures easier. It is most important that the implanting physician be aware of the danger and take steps necessary to avoid this potential complication.

A variation of the introducer technique involves retaining the guidewire. Instead of being removed together with the dilator, the guidewire is left in place so that the lead is passed through the sheath alongside it. The sheath is subsequently removed and peeled away. Occasionally, the size of the electrode (if the electrode is not isodiametric with the lead body) and the sheath precludes the passage of the lead alongside the guidewire. In this case, the guidewire is removed, the lead passed down the sheath, and the guidewire reinserted behind the electrode. This maneuver may succeed in this situation because it is the electrode that will not pass alongside the guidewire, whereas the lead body is thinner and leaves enough room in the sheath to accommodate the guidewire. Certain leads and sheath combinations are too tight to allow passage of both the electrode and guidewire. In this case, a larger sheath can be used. The retained guidewire may provide unlimited venous access and the ability to exchange or introduce additional electrodes by simply applying another sheath set to the guidewire. The retained guidewire should be held to the drape with a clamp to avoid inadvertent dislodgement.

The retained guidewire can be used as an intracardiac lead for recording of the atrial electrogram (EGM; to confirm atrial capture) or as an electrode for emergency pacing.

Occasionally, difficulty may be encountered with the standard-length sheath. Venous tortuosity, obstruction, and anomalies may preclude the advancement of the lead to the right side of the heart. This problem is usually solved by advancing a longer (e.g., 24 cm) sheath over the guidewire directly to the right atrium. This is also advisable when an extraction has been done during the procedure because of the weakening of the venous wall and/or partial venous obstruction ( Case Study 26-1 ). An appropriate length 0.035-inch guidewire is recommended. It is important that the guidewire be placed well distal to the problem area. Most long sheaths are prepackaged with a long guidewire. If an exchange is required, it is critical not to lose venous access. The original, short 0.035-inch guidewire should be retained and a standard 4-Fr sheath advanced over it. The dilator is removed, and a long 0.035-inch guidewire is passed down the 4-Fr sheath alongside the shorter guidewire. The 4-Fr sheath is removed, the short guidewire retained, and the long sheath passed over the long guidewire. Long sheaths can create their own challenges, however. Lead manipulation can be affected, and the advancement of these sheaths into the central venous circulation and right atrium engenders some risk, especially as they are advanced from the brachiocephalic veins into the SVC.

Methods of Multilead Introduction

The introducer approach is particularly useful in multilead systems. It has eliminated the earlier dilemma of needing to introduce multiple leads into a vein exposed by cutdown that may barely accommodate a single lead, with the resultant need for a second venous access site. For multilead systems, at least four methods that involve the introducer approach are described here. The first three can be used with any of the previously described percutaneous approaches.

Subcutaneous Defibrillation Electrodes

When the endocardial lead system alone fails to provide an adequate defibrillation threshold, a subcutaneous patch or a single-coil subcutaneous lead may be employed. A small electrode patch may be added through a small left anterior chest incision. This incision is usually placed along the left inframammary skinfold. A subcutaneous pocket is developed in the vicinity of the anterior axillary line. The patch is sutured to the chest wall, and the proximal lead is tunneled to the ICD pocket (see subsequent discussion on tunneling). A variation on this system is a subcutaneous coil, manufactured by Medtronic, that is implanted through a single slittable sheath. The sheath is first loaded onto a blunt-tipped, malleable rod that is hand-molded to slide subcutaneously from the prepectoral pocket around the lateral chest wall to the middle of the posterior chest wall at the level of the mid-left ventricle and medially 1 to 2 cm from the spine. The lead is inserted into the sheath after removal of the stylet, and then the sheath is slit for removal. As mentioned, the subcutaneous patch or the coil can be connected directly to the header or often is Y -connected with the SVC coil into the header with an adapter.

Azygos Vein Defibrillation Coil

As an alternative approach in patients with high defibrillation thresholds and routine endocardial lead systems, particularly in patients in whom implantation is performed from the right side, one can place a defibrillation coil (e.g., Medtronic Model 6937A) into the azygos/hemiazygos venous system. The advantages of use of the azygos vein defibrillation coil theoretically relates to the shift in the electrical defibrillation vector that occurs between the RV apex electrode and the azygos vein electrode, shunting the electric current posteriorly and more directly across the left ventricle. As with subcutaneous electrodes, this coil electrode can then be incorporated into the defibrillation system instead of using an SVC coil, or it can be coupled with an SVC coil by a Y adapter.

The azygos vein ascends in the posterior mediastinum to the level of the fourth thoracic vertebra, then arches anteriorly above the right pulmonary hilum to terminate in the SVC before the SVC penetrates the pericardium. The body of the azygos vein lies anterior to the thoracic vertebrae and to the right of the descending aorta, directly behind the heart. The technique for placing the azygos vein defibrillation coil involves cannulation of the ostium of the azygos vein in the posterior portion of the high SVC. This can be accomplished from either the right or the left axillary vein. A 6-Fr Judkins right 4 (JR4) diagnostic coronary catheter loaded with a 0.035-inch–diameter Wholey wire is advanced through a long, 45-cm, 9-Fr peel-away sheath with a hemostatic valve into the high SVC. The ostium of the azygos vein lies posteriorly and usually can be easily cannulated by probing with a combination of the JR4 catheter and the Wholey wire. The wire is advanced inferiorly to a position at or below the diaphragm, and the JR4 diagnostic catheter is advanced over the wire to the area of the diaphragm. In turn, the peel-away sheath can then be advanced over the combination of the wire and the inner catheter to approximately the same location at the level of the diaphragm. The wire and inner catheter are then removed from the sheath and replaced with the defibrillation coil lead, which is advanced through the sheath with the tip at approximately the level of the diaphragm ( Fig. 26-43 ). It can be difficult to advance the lead beyond the sheath, so it is important to try to position the tip of the sheath at the diaphragm. The sheath can then be removed and the azygos vein defibrillation coil used in place of the SVC coil port in the defibrillator. Venography can be used once the azygos vein os is cannulated, but this is not necessary in most cases. In our experience to date, there is no advantage of either the azygos or the hemiazygos over the other, as long as the tip of the lead is as near the diaphragm as possible.

Cardiac Implantable Electronic Devices Connectivity

Lead is important when considering a system upgrade. Preprocedural consideration of this is critical in obviating very challenging intraprocedural problems.

Occasionally, ipsilateral venous access is impossible. Either the vessel is thrombosed or some form of obstruction precludes the placement of additional leads from the same side. In this case, contralateral venous access can be achieved, and the leads are tunneled back to the original pocket ( Fig. 26-44 ). Early injection of radiographic contrast material may expedite the decision to use this approach. The use of the contralateral axillary or subclavian (rather than cephalic) vein is recommended for this approach. The distance to the original pocket is less, and the new lead is not as susceptible to dislodgement. The same percutaneous techniques and precautions as previously described are used for the venipuncture/introducer approach. The only difference is the size of the skin incision; only about 1 to 1.5 cm is required. The incision need only be large enough to allow anchoring of the lead and securing of the suture sleeve. As in an initial implantation, the incision should be carried down to the pectoral fascia. Once the lead has been positioned and secured, it can be tunneled to the original pocket.

The maneuver of passing a lead or catheter through tissue from one location to another is referred to as tunneling. It always involves the passage of a catheter from one wound through tissue to a second wound remote from the first. An example is the placement of a lead through the internal jugular vein. The lead is passed from the jugular incision through the tissue over (or under) the clavicle to the CIED pocket in the pectoral area. With the development of implantable defibrillator lead and patch systems that did not require thoracotomy, tunneling initially became a necessity. Current systems rarely require tunneling, though it is a useful tool even today.

A number of techniques are available for tunneling. They differ in level of trauma to the tissue and lead. As a rule, the least traumatic technique is desirable. A popular older technique is to place the proximal end of the lead or leads to be tunneled in a 14-inch Penrose drain ( Fig. 26-45A ). A gentle, nonconstricting tie is applied around the drain just distal to the lead connector (see Fig. 26-45B ). The track of the tunnel from the satellite wound to the pocket is infiltrated with local anesthesia by means of an 18-gauge spinal needle. The free end of the Penrose drain is then brought to the receiving wound from the satellite wound in the subcutaneous tissue. This can be accomplished with several techniques. The first technique involves the use of a Kelly clamp or uterine packing forceps. The tip of the clamp is pushed bluntly in the subcutaneous tissue from the receiving wound directly to the satellite wound. Care is taken to keep the tunnel as deep as possible, usually on the surface of the muscle. The free end of the Penrose drain is grasped and pulled back from the satellite wound to the receiving wound. The remainder of the Penrose drain containing the electrode connector pin is pulled through the track to the receiving wound. The tie is released, and the Penrose drain is removed ( Video 26-4 ).

A second, older technique delivers the Penrose drain to the receiving wound by use of a “passer,” usually a knitting needle or dilator. In this technique, the free end of the Penrose drain is fixed to the back end of the passer with a tie. The pointed tip of the passer is inserted into the satellite wound and pushed to the receiving wound. The tip of the passer is grasped and pulled into the receiving wound with the Penrose drain attached. The remainder of the Penrose drain with the lead is then pulled into the receiving wound.

A newer variation of this technique and more commonly used with today's single-connector CIED leads uses an introducer technique to establish the tunnel. After the track of the tunnel is infiltrated with an 18-gauge spinal needle, the needle is passed from the wound of origin to the receiving wound. A guidewire is passed through the needle into the receiving wound. A standard peel-away introducer large enough in diameter to accommodate the connector of the lead to be tunneled is then passed over the guidewire from the satellite incision to the receiving wound. The sheath can then be used to pass the lead, and the sheath is eventually removed and peeled.

The preceding techniques and principles (and creative variations) are used whenever tunneling is required. Even specialized tunneling tools are available for this and more general operative purposes. Tunneling with a clamp and directly grasping the lead should always be avoided because of the risk of damage to the lead.

Recently, venoplasty has been introduced as an alternative approach to venous access in situations of subtotal or total venous obstruction. This alternate approach is intended to avoid tunneling techniques or lead extraction as a means of gaining repeat venous access. Venoplasty is discussed in Chapter 32 , and the frequently associated need for transvenous lead extraction is discussed in Chapter 35 . These approaches have gained significant favor as they avoid both the need for tunneling and for involving other vascular beds.