Long-term Follow-up of Cardioverter-Defibrillator Implanted Under Conscious Sedation in Prepectoral Subfascial Position
Background Implantable cardioverter-defibrillators (ICDs) with intravenous electrode systems and downsized generators can be implanted by use of operative techniques similar to those employed for the insertion of permanent pacemakers. However, the safety, efficacy, and long-term follow-up of simplified implantation procedures remain to be evaluated. This report is a prospective long-term evaluation of nonselected patients receiving ICDs in the prepectoral subfascial position under conscious sedation.
Methods and Results Clinical characteristics of the 231 consecutive patients included a mean age of 63 years, a male-to-female ratio of 6.4, a left ventricular ejection fraction of 0.34, a mild-to-moderate heart failure in 91%, coronary artery disease in 84%, and a history of aborted sudden cardiac death or refractory ventricular tachyarrhythmias. Insertion of transvenous leads and prepectoral subfascial ICD implantation were performed in electrophysiology laboratories under local anesthesia and conscious sedation with intravenous midazolam and propofol. Successful implantation in all patients (operation time, 80±32 minutes, mean±SD) irrespective of body size and skin thickness was free of major complications, including need for emergency intubation. After surgery, 1 pocket hematoma, 1 seroma, and 1 pneumothorax required treatment. There was no operative or first-month mortality. During long-term follow-up averaging 453±296 days, six leads required repositioning, but pocket erosions or infections did not occur. First-year total survival was 97%.
Conclusions Implantation under conscious sedation of ICDs in the prepectoral subfascial position is a safe and effective procedure with low operative and postoperative morbidity and favorable long-term outcome.
Implantable cardioverter-defibrillators have traditionally been implanted in the surgical environment of the operating room, with cardiac surgeons performing the operation under general anesthesia. In recent years, the development of transvenous electrodes and downsized generators has permitted electrophysiologists to implant ICDs by use of techniques similar to those employed for the insertion of permanent pacemakers.1 2 3 4 5 6 Implantation of ICDs by cardiologists in electrophysiology laboratories has several advantages, such as support by a specialized staff and availability of appropriate monitoring and radiological equipment. However, there has been some reluctance to give up formal surgical support, including endotrachial intubation and general anesthesia.7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Plausible explanations for a conservative attitude are that patients receiving ICDs are often critically ill, that the devices remain bulkier and heavier than pacemakers, and that some form of deep sedation or general anesthesia is required for intraoperative testing of DFTs. Although reports from a few centers indicate the feasibility of implanting ICDs without general anesthesia,6 it remains to be demonstrated that such techniques applied systematically do not compromise safety and long-term outcome.
One unanswered question is whether pectoral implantation may lead to unduly frequent device migrations and pocket erosions.5 20 21 Like pacemakers, ICDs can be implanted in pockets fashioned in the prepectoral subcutaneous space. However, in lean or small subjects thought to be at risk of pocket erosion, formation of deep pockets beneath the major pectoral muscle has been recommended.5 20 21 Such pockets require deep dissection in planes occupied by important neurovascular structures. Beginning in mid-1993, we have prospectively evaluated implantation of ICDs in the prepectoral subfascial space, a procedure that minimizes trauma to subcutaneous tissue and avoids deep subpectoral dissection. Here, we report the long-term follow-up of 231 consecutive patients receiving ICDs in this location.
Patients of either sex >15 years old were eligible if they had survived at least one episode of aborted sudden cardiac death due to a ventricular tachyarrhythmia or had sustained hemodynamically unstable VT refractory to medical therapy. Patients within 7 days of an acute myocardial infarction and patients receiving a replacement ICD were excluded. Between the fall of 1993 and the spring of 1996 (a 30.3-month period), 231 consecutive patients fulfilling the inclusion and exclusion criteria were nonselectively entered into the study at two Texas medical centers. All patients had received coronary arteriography and an estimation of left ventricular ejection fraction either by contrast angiography, radionuclide ventriculography, or echocardiography. Baseline electrophysiology testing was performed in all patients by standard techniques.22 The institutional review boards of the participating institutions approved the study protocols. Written informed consent was obtained from all patients.
ICDs suitable for pectoral implantation included Jewel models 7219 D (n=164) and 7219 C (unipolar can system; n=41) from Medtronic; Cadet models V-115 D (n=6) and can model V-115 AC (n=7) from Ventritex; and Ventak Mini models 1740 (n=2) and 1741 (n=11) from Cardiac Pacemaker Inc. The weight and volume of the devices ranged between 129 and 139 g and 73 and 89 cm3.
For Jewel models, tripolar RV leads (Transvene-RV model 6936) in combination with unipolar SVC leads (Transvene-SVC model 6933) or, rarely, subcutaneous patch electrodes (model 6999) were used. For the Cadet models, RV and SVC leads (models RV02 and SV02) were inserted. For the Ventak Mini models, dual coil electrodes (Endotak DPS 0125) were used.
Before implantation, a radial arterial catheter was inserted and a finger pulse oximeter was attached to monitor arterial pressure and oxygenation. Medications, including prophylactic antibiotics (vancomycin, 15 mg/kg; gentamicin, 1 mg/kg), were given via a peripheral vein, and no central venous lines were placed so as to minimize the risk of defibrillator lead contamination. Monitored anesthesia care sedation was administered in the presence of an anesthesiologist. Dosages of premedication with midazolam (Versed, Roche Laboratories; 15 to 30 μg/kg IV) and continuous infusion of propofol (Diprivan, Stuart Pharmaceuticals; initial rate, 25 to 50 μg·kg−1·min−1 IV) were adjusted to achieve conscious sedation. Propofol was titrated throughout the procedure to ensure patient comfort without inducing loss of consciousness. Levels of consciousness and awareness were monitored by frequent interrogation (every 1 to 2 minutes) of the patients. Supplemental oxygen was delivered by nasal cannula or mask. For the induction of fibrillation and DFT testing, a bolus of propofol (400 to 500 μg/kg) was given to produce deep sedation. The efficacy of the propofol bolus, only 20% to 25% of a usual induction dose, reflected the potentiating effects of the premedication with midazolam and low-dose propofol.23 24 The aim was to produce a brief loss of responsiveness to glabellar tap and loud auditory stimulation (Ramsey score 5 to 625 or Adenbrooke's score 4 to 526 ). After recovery from deep sedation, patients were interrogated to assess amnesia for operative and postoperative events.
Transvenous electrodes were introduced into the venous system by transcutaneous cannulation of the left subclavian vein or by left cephalic vein cutdown. Under fluoroscopic control, RV electrodes were inserted into the RV apex, and vena cava defibrillation leads were advanced to the SVC. The final localization of the electrodes within the venae cavae was adjusted to obtain optimal DFTs. Use of unipolar can systems or of leads with dual defibrillation coils (Endotak) obviated the need for vena cava leads. Redundant extravascular segments of the electrodes were formed into loops stabilized with three suture sleeves to the underlying pectoral muscle. In a few patients (n=15), subcutaneous patch electrodes were required. After satisfactory R-wave sensing (>5 mV) and pacing thresholds (<1.0 V at 0.5-ms pulse width) had been demonstrated, the patients were prepared for DFT testing.
Methods for the induction of VF depended on the ICD model used and included low-energy shock on T wave (Jewel), 50-Hz AC current-burst (Jewel), and programmable rapid-burst pacing (Cadet, Ventak Mini). For the determination of DFTs, a step-down testing scheme minimizing the number of VF inductions was used.6 The requirements of a minimal 10-J margin of safety between the DFT and maximum output of the generators was met in all cases. If SVC lead repositioning and lead polarity reversal were unsuccessful in achieving defibrillation with the desired safety margin, a subcutaneous patch electrode was inserted. In patients receiving the active can system, defibrillation testing before device implantation was carried out with a titanium shell emulator. In all patients, lead position was verified and the device was retested before discharge, usually 1 or 2 days after implantation.
Under local anesthesia with 1% lidocaine, an infraclavicular incision ≈7 to 8 cm long was made. A pocket was fashioned in the prepectoral subfascial space, dissecting the variably developed areolar connective tissue between the deep layer of the superficial pectoral fascia and the superficial layer of the deep pectoral fascia. The device was anchored with sutures to the pectoral muscle and deep fascia. The skin was closed in three layers.
After implantation, the patients were first observed in the electrophysiology laboratory and then, if necessary, in holding facilities. Subsequently, they were sent to their rooms and received analgesics for pain and discomfort.
Follow-up care consisted of a wound examination at 1 week and a visit at 1 and 3 months, when chronic DFTs were tested. Subsequent visits with routine ICD interrogation were at 3-month intervals, but additional visits were scheduled when patients reported ICD shocks or other ICD-related events. At each visit, patients were asked about concerns related to body image, discomfort in the area of the implant, and impairment of shoulder motion.
Age, sex, primary indication, underlying cardiovascular disease, NYHA functional classification, and ejection fraction were characterized by absolute (count) and relative (percent )frequencies. Confidence limits (95%) for sample means were calculated as the sample mean±[t×SEM], where t is the appropriate critical value from the t distribution. If not otherwise specified, values expressed represent mean±SD. Survival curves were constructed according to the Kaplan-Meier (product-limit) method.
Clinical Characteristics of Patients
A total of 231 consecutive patients were treated with implantation of ICDs from three manufacturers. The mean age was 63 years (range, 17 to 85 years), and the majority (87%) were men (Table 1⇓). The body mass index (desirable range, 20 to 25 kg/m2) was above the overweight limit for women (>27.3 kg/m2) but not for men (<27.8 kg/m2).27 According to the HANES III data (a US nutrition survey, 1988 to 1991), the estimated prevalence of obesity is 34% for the general US population and 52% for the black female population.27 In 84% of the patients, coronary artery disease documented by arteriography was the major underlying disease. Other diagnoses included idiopathic cardiomyopathy, valvular heart disease, and primary electrical disease, including long-QT syndrome. The mean left ventricular ejection fraction was 0.34. Functional classification of the patients according to the NYHA yielded the following percentile distribution: class I, 34%; class II, 57%; class III, 8%; and class IV, 1% (2 patients) (Table 1⇓).
The primary indication for ICD implantation was aborted sudden cardiac death in 92 patients (40%), monomorphic VT without history of resuscitation in 125 (54%), and nonsustained VT with syncope and inducible VT at electrophysiological study in 14 (6%).
The approach was through the subclavian vein in 126 patients and through the cephalic vein in 105. Of the 231 patients, 155 (67%) received an RV/SVC configuration, 48 (21%) a unipolar can configuration, 15 (6.4%) a subcutaneous patch, and 13 (5.6%) dual-coil leads. RV leads were positioned in 229 patients in the RV apex, but in 2, anchoring to the RV septum was required to meet desired thresholds of pacing and sensing.
Fibrillation Induction and DFTs
Induction was successful using T-wave shocks in 185 patients (80%), 50-Hz bursts in 20 (9%), and manual bursts in 26 (11%). The operational DFT securing a 10-J safety margin averaged 17±5 J. The average number of ventricular fibrillation inductions per patient was 3.98±2.50 (95% CI, 3.85 to 4.11). The mean defibrillation lead impedance was 62±10 Ω. In two patients set up to receive a unipolar Jewel 7219C, the criteria of defibrillation were not met at implantation, and patients were given a Jewel 7219D instead.
In all patients, prepectoral subfascial implantation was performed without difficulty or operative complications. The short action of propofol permitted easy and rapidly adjustable titration of the depth of sedation. During conscious sedation, there were no episodes of sustained arterial hypotension (systolic pressure <90 mm Hg) or apnea requiring breathing assistance or emergency intubation. Similarly, no instance of mechanical breathing assistance was required during deep sedation. The length of the procedure measured from the time of the first injection of lidocaine to the time of completed skin closure averaged 80±32 minutes (range, 21 to 225 minutes). The vast majority of the patients recovered within 5 to 7 minutes (drop of Ramsey sedation score from 5 or 6 to <325 ) after termination of DFT testing. Delayed awakening (>14 minutes) was observed in 4 elderly (>75 years) male patients. After recovery from sedation, interrogation of patients in no instance revealed recall of operative interventions performed under deep sedation.
Arrhythmic episodes and therapy delivered
During follow-up, 94 patients (41%) received appropriate ICD treatment for ventricular tachyarrhythmias. Fifty-three patients (23%) received shocks only; 12 (5%), antitachycardia pacing only; and 29 (13%), combinations of these therapies.
Follow-up exceeded 1 year in 117 patients (51%) and 2 years in 65 patients (28%). The follow-up for all patients averaged 453±296 days (median, 368 days; range, 24 to 924 days). At the first two visits after implantation, 17 patients complained about discomfort in the area of the implant. Subsequently, however, patients appeared to develop tolerance for the devices, and persistent complaints were not recorded. Review of the records of patients in the lowest quintile (n=46) of body mass index (values <20 kg/m2 in 8 patients) revealed three complaints or concerns about conspicuousness of the devices, whereas only one such complaint was registered in the rest of the patients.
After surgery, 3 pocket hematomas with 1 requiring aspiration, 1 seroma over a subcutaneous patch, 1 thrombosis of the SVC, and 1 pneumothorax were recorded. All resolved without sequelae. During follow-up, 6 lead repositionings were required, but lead fractures did not occur. Two devices were explanted in patients undergoing cardiac transplantation because of intractable heart failure. No patient suffered RV perforation or thromboembolic complications. Not a single case of pocket erosion or infection was recorded during the entire follow-up.
There were no in-hospital deaths and no deaths within 30 days of implantation. Eleven deaths were recorded during the entire follow-up, 5 of which were noncardiac (2 chronic lung diseases with respiratory failure, 1 lung carcinoma, 1 after coronary bypass operation, and 1 aortic aneurysm with intestinal infarction) (Figure⇓). Among the 6 cardiac deaths, 5 were nonsudden (4 heart failures, 1 acute myocardial infarction with cardiogenic shock) and 1 sudden. Total mortality at 1 year was 4 deaths, 3 noncardiac and 1 sudden cardiac. First-year total survival was thus 97% (Figure⇓).
The development of reliable transvenous lead systems and downsized generators has greatly facilitated the implantation of ICDs.1 2 3 4 5 6 28 29 However, optimal techniques for the implantation of these devices are controversial, and it remains unclear whether implantation should be performed in operating rooms or electrophysiology laboratories, whether operators should be surgeons or electrophysiologists, whether local or general anesthesia should be administered, and whether implantation should be in the abdominal, subpectoral, or prepectoral position.1 5 19 20 21
In an early report on pectoral ICD implantation under general anesthesia, the authors concluded that consistent subcutaneous prepectoral implantation was possible, but the study was limited to 13 patients, and no follow-up was provided.3 In two recent reports, nonthoracotomy devices were implanted in the abdominal position, but the reason for this preference was not explained,6 14 and one group admitted that frequent lead complications may have been related to operative technique.14 In a recent international study, a variety of implantation procedures were analyzed only in terms of epicardial or endocardial lead placement, with no information about anesthetic and surgical techniques or pocket localization.17
In the present study, ICDs were implanted in the prepectoral subfascial position by electrophysiologists who had worked initially as a single team and had adopted the same anesthetic and operative techniques. All patients, regardless of body mass index, sex, and pectoral skinfold thickness, received the devices in this position. The implantation protocol, including conscious sedation with low-dose midazolam and propofol, was implemented in all patients irrespective of pulmonary function status. Implantations were performed in the presence of an anesthesiologist, but in no case was there a need for emergency intubation and mechanical ventilation. This report differs from multicenter series in which operators were allowed to enroll nonconsecutive cases selectively in a protocol.17 To the best of our knowledge, the only report of ICD implantation without general anesthesia in consecutive patients (n=27) involved abdominal generator insertions.6 The cost-effectiveness of ICDs operating through transvenous lead systems compared with the previous thoracotomy devices has recently been emphasized.30 31 Elimination of inhalation anesthesia is an additional factor that should help to reduce the cost of ICD implantation.
Patients with potentially lethal arrhythmias entered into this study are probably similar to patients who received ICD treatment in other American referral centers. The clinical characteristics of our patients resemble those of other published studies. In Table 2⇓, we have calculated mean values for patient characteristics provided by 18 studies cited in this article.2 3 6 7 8 9 10 11 12 13 14 15 17 18 23 24 25 26 There were modest differences in the NYHA functional classification between our study and the summarized reports, although left ventricular ejection fractions were very similar (34% versus 33% in studies reviewed). The pronounced variances in NYHA classification encountered in the studies reviewed (95% CI, 6% to 20% for class III, Table 2⇓) may partly reflect the limitation of categorizations based on subjective interpretations of patient histories elicited. It is clear that neither in our nor in the other reports did severe heart failure patients represent a prominent group. The similarity of the data in our report to those of the combined studies suggests that our patient population was not distinguished by special features that might explain favorable outcome measures such as absence of postoperative and perioperative mortality, a total first-year survival rate of 97%, and an absence of lead fractures and pocket erosions.
There appears to be no consensus regarding the optimal pocket localization for currently available ICDs.5 19 20 21 Surgeons performing augmentation mammoplasties have for many years evaluated the insertion of foreign bodies in pockets created in the prepectoral (subglandular) or subpectoral (submuscular, submusculofascial) spaces.32 33 A consensus has developed that compared with submuscular insertion, prepectoral implantation is more likely to result in periprosthetic capsular contracture and implant immobilization, a potentially desirable outcome in the case of ICDs.32 33 Surgeons placing foreign bodies in the pectoral position should be conversant with the different planes defined by the pectoral fascial system.32 Recent reports on pectoral ICD implants designate pocket localizations as subcutaneous (prepectoral) or subpectoral, but the plane entered in the subcutaneous-prepectoral approach is not explicitly defined.3 5 Accordingly, it is unclear whether the devices are implanted partly within the subcutaneous fatty tissue or underneath it. In the present study, the prepectoral space (also called retromammary space) was entered in a plane delimited by the deep layer of the superficial pectoral fascia and the superficial layer of the deep pectoral fascia.32 The two fascial layers are loosely attached to each other by variously developed areolar connective tissue. Pockets in the retromammary space have theoretical advantages compared with other pectoral implantation sites. First, the skin overlying the implant is protected by an interposed layer of connective tissue. Second, this localization minimizes foreign body exposure to subcutaneous fatty tissue, the trauma of which leads to release of proinflammatory lipolytic products34 and so-called fat necrosis (traumatic panniculitis35 ). Third, compared with subpectoral implantation, this localization obviates dissection along important neurovascular structures and may help preserve pectoral muscle integrity and shoulder motion.20 21 Last, the superficial localization facilitates generator and lead revisions. Our series fails to confirm that prepectoral subfascial implantation is likely to be complicated by pocket erosion. This potential complication is often used as an argument to justify operations requiring deep subpectoral dissection.20 21
In summary, we report for the first time in consecutive nonselected patients a long-term follow-up of ICDs implanted with operative methods similar to those used for the insertion of permanent pacemakers. Our results compare favorably with those of other studies,8 10 11 14 17 and there is no indication that our simplified implantation method, including avoidance of inhalation anesthesia, exposes patients to increased risks.
Selected Abbreviations and Acronyms
|SVC||=||superior vena cava|
Guest editor for this article was Hein J.J. Wellens, MD, University Hospital Maastricht, Maastricht, the Netherlands.
- Received July 23, 1996.
- Revision received October 7, 1996.
- Accepted October 14, 1996.
- Copyright © 1997 by American Heart Association
Fitzpatrick AP, Lesh MD, Epstein LM, Lee RJ, Siu A, Merrick S, Griffin JC, Scheinman MM. Electrophysiological laboratory, electrophysiologist-implanted nonthoracotomy-implantable cardioverter/defibrillators. Circulation. 1994;89:2503-2508.
Bardy GH, Raitt MH, Jones GK. Unipolar defibrillation systems. In: Singer I, ed. Implantable Cardioverter-Defibrillator. Armonk, NY: Futura Publishing Co; 1994:365-376.
Kleman JM, Castle LW, Kidwell GA, Maloney JD, Morant VA, Trohman RG, Wilkoff BL, McCarthy PM, Pinski SL. Nonthoracotomy- versus thoracotomy-implantable defibrillators: intention-to-treat comparison of clinical outcomes. Circulation. 1994;90:2833-2842.
Venditti FJ Jr, Martin DT, Vassolas G, Bowen S. Rise in chronic defibrillation threshold in nonthoracotomy implantable defibrillators. Circulation. 1994;89:216-223.
Strickberger SA, Hummel JD, Daoud E, Niebauer M, Williamson BD, Man KC, Horwood L, Schmittou A, Kalbfleisch SJ, Langberg JJ, Morady F. Implantation by electrophysiologists of 100 consecutive cardioverter-defibrillators with nonthoracotomy lead systems. Circulation. 1994;90:868-872.
Schwartzman D, Nallamothu N, Callans DJ, Preminger MW, Gottlieb CD, Marchlinski FE. Postoperative lead-related complications in patients with nonthoracotomy defibrillation lead systems. J Am Cardiol. 1995;26:776-786.
Wever EFD, Hauer RNW, van Capelle FGL, Tijssen JGP, Crijns HJGM, Algra A, Wiesfeld ACP, Bakker PFA, Robles de Medina EO. Randomized study of implantable defibrillator as a first-choice therapy versus conventional strategy in postinfarct sudden death survivors. Circulation. 1995;91:2195-2203.
Zipes DP. Are implantable cardioverter-defibrillators better than conventional antiarrhythmic drugs for survivors of cardiac arrest? Circulation. 1995;91:2115-2117.
Zipes DP, Roberts D, for the Pacemaker Cardioverter-Defibrillator Investigators. Results of the international study of the implantable pacemaker cardioverter-defibrillator: comparison of epicardial and endocardial lead systems. Circulation. 1995;92:59-65.
Villacastin J, Almendral J, Arenal A, Albertos J, Ormaetxe J, Peinado R, Bueno H, Merino JL, Pastor A, Medina O, Tercedor L, Jimenez F, Delcan JL. Incidence and clinical significance of multiple consecutive, appropriate high-energy discharges in patients with implanted cardioverter-defibrillators. Circulation. 1996;93:753-762.
Epstein AE, Kay GN. Implantation of cardioverter-defibrillators in the electrophysiology laboratory. In: Singer I, ed. Implantable Cardioverter-Defibrillator. Armonk, NY: Futura Publishing Co; 1994:357-364.
Ramsey MAE, Savage TM, Simpson BRJ, Goodwin R. Controlled sedation with alphaxalone-alphadalone. Br Med J. 1974;2:656-659.
Russell CM, Williamson DF, Byers T. Can the year 2000 objective for reducing overweight in the United States be reached? A simulation study of the required changes in body weight. Int J Obesity. 1995;19:149-153.
Hammill SC, Packer DL, Stanton MS, Fetter J, the Multicenter PCD Investigator Group. Termination and acceleration of ventricular tachycardia with autodecremental pacing, burst pacing, and cardioversion in patients with an implantable cardioverter-defibrillator. Pacing Clin Electrophysiol. 1995;18:3-10.
Swartz JF, Fletcher RD, Karasik PE. Optimization of biphasic waveforms for human nonthoracotomy defibrillation. Circulation. 1993;88:2646-2654.
Bostwick J III. Plastic and Reconstructive Breast Surgery. St Louis, Mo: Quality Medical Publishing Inc; 1990:57-97.
Spear SL, Dawson KL. Augmentation mammoplasty. In: Cohen M, ed. Mastery of Plastic Surgery. Boston, Mass: Little, Brown, & Co; 1994:2099-2113.
Su WPD. Diseases of the subcutaneous tissue. In: Moschella SL, Hurley HJ, eds. Dermatology. Philadelphia, Pa: WB Saunders Co; 1992:1312-1332.