HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): J. Ernesto Molina Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Molina, J. E. Articles by Adler, S. Search for Related Content PubMed PubMed Citation Articles by Molina, J. E. Articles by Adler, S.

J Thorac Cardiovasc Surg 1995;110:258-264
© 1995 Mosby, Inc.

SURGERY FOR ACQUIRED HEART DISEASE

Crinkling of epicardial defibrillator patchesA common and serious problem

Minneapolis, Minn.

From the Departments of Surgery and Cardiology, University of Minnesota, Minneapolis, Minn.

Received for publication June 15, 1994. Accepted for publication Nov. 15, 1994. Address for reprints: J. Ernesto Molina, MD, University of Minnesota, Box 182, 420 Delaware St., S.E., Minneapolis, MN 55455.

Abstract

The durability and reliability of the implantable cardioverter-defibrillator epicardial patch systems have not been reported. In 128 consecutive patients such systems manufactured by Cardiac Pacemakers, Inc. (St. Paul, Minn.) or Medtronic, Inc. (Minneapolis, Minn.) were implanted with 100% follow-up to investigate the rate of patch crinkling and its consequences. A total of 122 patients survived the operation (operative mortality, 6 patients; 4.7%). Ninety-four patients received Cardiac Pacemakers, Inc. AICD patches and 28 received Medtronic PCD patches. Patients had chest x-ray studies every 3 to 6 months and function of the defibrillator was checked every 3 months. Late mortality occurred in 17 patients (13%) leaving a total of 105 long-term survivors (82%) to the present. Among 122 survivors, severe crinkling of the patches occurred in 48 patients—33 in the Cardiac Pacemakers, Inc. AICD group (36%) and 15 in the PCD group (54%)—within 2 years of the implant. Crinkling of patches caused not only malfunction of the system, but also cardiac pain in three patients. Crinkling occurred as early as 2 months after implant and progressed throughout the period of observation. Fourteen patients later required implant of an additional transvenous defibrillator because of failure of the epicardial system. The percentage of transvenous implantable cardioverter-defibrillator systems needed was higher for the Medtronic group (28%) than for the Cardiac Pacemakers, Inc. AICD group (6.3%). Both systems have shown an unacceptably high rate of patch crinkling that occurs in a relatively short time. There is no difference whether a thoracotomy or midline sternotomy is used or whether the patches are implanted intrapericardially or extrapericardially. The quest should continue for a better patch system design. (J THORAC CARDIOVASC SURG1995;110:258-64)

The use of implantable cardioverter-defibrillators (ICDs) has become the treatment of choice for life-threatening ventricular tachyarrhythmias. From 1982 until late 1991, the standard implantable system entailed the use of epicardial patches that had to be positioned around the heart. The electrical discharge was delivered through these patches. All the implantable units were manufactured by Cardiac Pacemakers, Inc. (CPI) (St. Paul, Minn.). During the 1980s the CPI system consisted of semirigid patches of a titanium mesh embedded in a clear plastic square design. This design has remained unchanged and thus provides for a consistent, long follow-up of the fate of these patches for more than 10 years.

In 1990, the Medtronic PCD unit (Medtronic, Inc, Minneapolis, Minn.) became available with a different type of patch. Most reports on the durability and stability of the defibrillation threshold have looked at the CPI-AICD system ^1-6 and have mainly assessed its function. ^5-7 The report by Goodman and associates ⁸ is the only one in the literature on the fate of the defibrillator patches that showed major crinkling: a complication that is noted only if follow-up pays attention to it over a sufficient period of time. Other reports of isolated cases have indicated a relatively low crinkling rate, ^5,6,9,10 but none has investigated the true frequency of crinkling beyond anecdotal cases. ^5,11 Moreover, no reports have looked into the crinkling rate of the Medtronic patch system.

Crinkling in epicardial patches eventually will affect the function of the ICD system and require further surgical intervention.

In 1987 our institution started implanting AICD systems manufactured by CPI; since 1989 we have also used the PCD Medtronic system. Because of the strict and meticulous follow-up that our patients must have, we noticed that a significant number of the epicardial patches showed crinkling on radiographic examination. The defibrillation thresholds sometimes changed to the point that the system no longer functioned, prompting either the addition of new leads or the implant of an entire new system. Because more epicardial patches have been implanted since the Medtronic system was released for general use, we decided to investigate how reliable and safe the patch designs are.

PATIENTS AND METHODS

Since 1987, 128 patients have received an epicardial ICD implant system at the University of Minnesota for either ventricular tachycardia (n = 68) or ventricular fibrillation sudden death (n = 60). We used patches from two manufacturers: 99 patients received the CPI-AICD and 29 the Medtronic PCD.

To implant these patches we used a left thoracotomy in 95 patients and a midline sternotomy in 33 patients. Of the patients undergoing midline sternotomy, nine had other procedures done simultaneously: six coronary artery bypass operation and three mitral valve operation. In the coronary artery bypass group, three patients underwent implant of the patch only at the time of the aorta-coronary bypass operation; these patients returned later to have the pulse generator implanted and the entire system tested. All three patients who underwent mitral valve operation had the patches implanted at the same time; all three returned later to have the pulse generator implanted.

The operative mortality rate was 47% (n = 6). A total of 122 survivors were followed up for 3 months to 6 years.

Assessment of the degree of crinkling
The patches as supplied by the company are flat and straight, but as they are applied on the surface of the heart they adapt to follow the contour of the organ. Therefore a mild degree of curving or folding is accepted as normal. Usually we considered the patch normal as long as the curvature was not less than 160 degrees and followed the outline of the heart. Mild crinkling was defined as a curvature between 160 and 135 degrees, a moderate crinkling was curvature between 90 and 135 degrees, and severe crinkling was any curvature sharper than a 90-degree angle.

Crinkling occurred on various portions of the patch throughout follow-up. We were usually not concerned with the function of the system unless crinkling was severe, in which case we monitored the system more often (every 6 months) until the patch remained unchanged for 12 months and then every 12 months.

Follow-up
We obtained 100% follow-up by contacting patients directly by phone or mail or through their private physicians or cardiologists.

Once the patient was contacted, chest x-ray films were obtained and the degree of patch crinkling assessed. We then made recommendations to the patients to repeat the roentgenography at certain intervals, depending on the degree of crinkling: the interval for severe crinkling was every 6 months; for mild or moderate crinkling it was every 12 months. All x-ray films were reviewed at the University of Minnesota to ensure uniform comparison. Five patients did not receive the defibrillator at the University of Minnesota, but were being followed up by us and were thus entered into this study.

Technique of implantation
The 128 patients underwent either a left thoracotomy or a midline sternotomy. The patches were secured to the pericardium with interrupted stitches. The number of stitches varied from six to eight, and they were usually placed at the edges of the patch to the pericardium once the proper position was achieved. In the two-patch system, a large size patch over the left ventricle was positioned lengthwise along the left border of the heart, with the cardiac border in the middle of the patch. The connecting end usually was oriented toward the apex, but in a few cases was oriented toward the base of the heart. The anterior patch was usually attached with six or more stitches to the pericardium overlying the right ventricle, with the length of the patch oriented from the right border of the heart toward the pulmonary outflow tract.

The pericardium was closed with a few interrupted stitches. Enough space was left for the pericardial fluid to drain into the pleural cavity where the chest tube drainage system was implanted. The sensing electrodes were placed over the left ventricle. We exclusively used the sutureless screw-type design (CPI model 4312, Medtronic model 6917).

All leads were tunneled down along the rectus muscle sheath into the left flank of the abdomen where a pocket was made in the subcutaneous tissue.

Patches studied
We studied only the two systems manufactured by CPI and Medtronic because they were the only types implanted at the University of Minnesota between 1987 and 1994. In addition, four patients were added to our follow-up group who had not originally undergone implantation at our institution: we had the original reports on these patients and all had received the CPI system.

A total of 29 patients received the Medtronic system: 13 received a two-patch system and 16 received a three-patch system (Table I). When three patches were used, the third patch was implanted on the inferior surface of the heart over the dome of the diaphragm: the left patch over the left ventricle was moved more anteriorly over the anterolateral wall of the left ventricle, and the right patch remained in the same position over the right anterior wall of the right ventricle.

Length of follow-up
Because the CPI system had been used for a longer period (1987 to 1994), the length of follow-up extended to 6 years. For the Medtronic system (1989 to 1994), the longest follow-up was 4 years. The mean follow-up for both systems combined was 3.8 years.

Mortality and morbidity
Within 6 weeks after the operation, six patients (4.7%) died, all of heart failure. The ejection fraction ranged from 0.14 to 0.20 (estimated by preoperative two-dimensional echocardiographic examination). In fact, two patients needed inotropic drug infusion during operation to maintain adequate blood pressure after they underwent from four to seven induced episodes of ventricular fibrillation and defibrillation with use of the external system.

Other complicating factors in these six patients were renal failure, respiratory insufficiency, and inability to be weaned off the respirator after general anesthesia. In three of them, the isoenzyme creatine kinase MB fraction was elevated, which suggests some myocardial damage during the testing.

Of the 122 survivors, 17 later died during follow-up (13%), leaving a total of 105 long-term survivors (82%). Of the 17 late deaths, three were not witnessed and the cause is unknown. Failure of the defibrillator system was suspected in two of these patients because the threshold had changed throughout follow-up. In two of these patients, the defibrillator system was removed at their request with full knowledge of the risks involved; both died 3 months later, probably of arrhythmia.

Of the 128 total patients, 18 had complications after the operation, including three who had infections that necessitated removal of the ICD system. Constrictive pericarditis occurred in two patients, whose patches were removed and converted to a transvenous system with a chest-wall patch. Hematomas developed in another two patients. Two patients underwent reexploration because of bleeding, which in one case was caused by dehiscence of the patch from the cardiac surface. Another patient had a pulmonary embolism, but recovered.

Crinkling of epicardial patches
The degree of crinkling shown on chest x-ray films before radiographic examination was analyzed independently for the PCD and AICD systems. We classed 48 patients as having severe crinkling: 15 with the PCD and 33 with the AICD. Moderate crinkling was determined in 10 patients. Table II shows the distribution of crinkling for the two systems tested.

View larger version (108K): [in this window] [in a new window]   Fig. 1. Crinkling of CPI-AICD patch system at 2 years. A, Posteroanterior view. Outline of patch can be seen on left border of heart. B, Lateral view showing severe degree of crinkling of posterolateral patch.

View larger version (112K): [in this window] [in a new window]   Fig. 2. Gradual crinkling of PCD patch system. A, Lateral view at time of implant. B, Lateral view of same patient 9 months later. Severe crinkling of both patches can be observed.

Placing the patch extrapericardially, as was done for some patients, did not solve the crinkling problem. Fig. 3 shows a posterior patch implanted extrapericardially that had the same crinkling rate as its intrapericardial counterparts. As follow-up lengthened, the patch not only crinkled more, but also migrated through the chest cavity.

View larger version (168K): [in this window] [in a new window]   Fig. 3. Lateral view at 1 year after operation in patient who underwent implant of PCD patch system extrapericardially. Extrapericardial posterior patch has crinkled and migrated across chest cavity toward the back.

Implant of transvenous ICD
In 14 patients, because of the late failure of the patch system (Table III) to properly defibrillate the heart, we implanted a new transvenous (nonthoracotomy) lead. All the transvenous systems used at our institution were Transvene-PCD (Medtronic) types. Table IV displays the number of patients who eventually required a transvenous system because of malfunctioning of the epicardial type. The percentage was significantly higher for the Medtronic PCD system (28%) versus the CPI-AICD system (6.3%).

A 24-year-old woman with Marfan's syndrome previously had an epicardial Medtronic patch system; severe mitral regurgitation developed for which mitral valve replacement was recommended. She had frequently reported cardiac pain that occurred unpredictably, day or night, and was not related to any specific activity. Severe crinkling of her PCD patches had developed but the unit was still working. Because she was young and the coronary arteriogram was perfectly normal, we concluded that the cause of the pain was probably the crinkled posterior patch located between the inferior surface of the heart and the diaphragm. At the time of mitral valve replacement, we removed this patch. After operation the pain completely disappeared.

A 48-year-old man had patch crinkling in the same position as the woman just described. He also needed the patch removed because of infection. As soon as the patch was removed through a left thoracotomy, the pain disappeared.

The third patient was a heart transplant recipient whose ICD system was implanted while he was on the organ waiting list. He reported cardiac pains that were positional and distinct from the angina. His pain disappeared after the heart transplant.

These three cases point to the fact that crinkling can cause not only inadequate function, but also actual injury to the heart (because of the rigidity of this foreign body pressing on the ventricles).

As of May 1, 1994, 35 patients who still have the original, but crinkled, defibrillator patch system are alive with the system adequately working. These patients are now being followed up periodically for function and for worsening of the patch crinkling (which may become more severe and may also lead to migration within the chest cavity).

DISCUSSION

The epicardial patch system for defibrillators is being replaced these days by the transvenous approach, which has less morbidity and mortality. However, the epicardial system will always be needed for certain patients, ^12,13 particularly for those in whom adequate defibrillation thresholds cannot be reached with the transvenous approach, for children, for patients with occluded venous systems, for patients receiving dialysis, or for patients with transvenous devices already implanted in the upper veins of the body.

The two most common ICD epicardial patch systems now on the market, namely CPI and Medtronic, have shown an unacceptable high rate of patch crinkling, which occurs in a relatively short period. In our study, the mean value was 1 year and 8 months. With the Food and Drug Administration's release of the Medtronic epicardial patch systems, more will be implanted in the future. Thus it is likely that the number of patients with crinkled patches will soon increase. Not only will the patch eventually fail, but it also may migrate and erode into nearby organs or cause pain of cardiac origin. If an intrapericardial implant technique is used, compression will probably be confined to the heart, but in the extrapericardial system the patches may migrate into the chest cavity or mediastinum, as we have seen.

When some of these crinkled patches were surgically removed, the findings varied; for example, the suture material might have pulled through the pericardium or the pericardium might follow the contour of the crinkled patch. In two patients with Medtronic system patches we found that stitches had pulled through the patch material itself. Because crinkling and folding of the patches has also been observed, in cases in which the patches were placed in the subcutaneous tissue of the chest, in conjunction with implant of transvenous leads, we changed our technique to sewing the patches with a running nonabsorbable suture, usually monofilament material, along the entire circumference of the patch. However, this technique cannot be followed up in most of the cases when patches are placed in the pericardium because of the lack of sufficient exposure. Actually, even placing six interrupted stitches in various areas of the patch to anchor it to the pericardium is something quite difficult. Therefore most of the free edges of the patch are not anchored.

Once a defibrillator system becomes ineffective, a new system is required, usually a transvenous device. Although better, transvenous devices are not free of complications either. ¹² Complications we have already seen include vein thrombosis of the subclavian veins extending into the arm and severe fibrotic obstruction of the innominate veins or superior vena cava (which also makes heart transplantation for those patients rather difficult and more risky). Perforation of the right ventricle by the heavier and stiffer transvenous defibrillator leads has caused cardiac tamponade, placing patients at high risk, particularly those with poor ventricular function. Infections with endocarditis have also been caused by these intravascular devices. Migration of the floating leads in the superior vena cava or innominate vein is also a problem.

Thus the need remains for a better epicardial patch system that will allow lower thresholds for effective defibrillation and will permit smaller pulse generators to be implanted in the chest. The new Medtronic Jewel unit, for example, is one such system now in clinical trials.

Surgeons and cardiologists must be fully aware of all these problems. In cooperation with the medical device industry, they should make an effort to design safer systems to treat patients who already have marginal cardiac reserves.

References

1. Tchou PJ, Kadri N, Anderson J, et al. Automatic implantable cardioverter defibrillators and survival of patients with left ventricular arrhythmias. Ann Intern Med 1988;109:529-34.
   
2. Winkle RA, Mead RH, Ruder MA. Long-term outcome with the automatic implantable cardioverter defibrillator. J Am Coll Cardiol 1989;13:1353-61.[Abstract]
   
3. Nissan S, Mower MM, Thomas A, et al. Patient survival comparison in three generations of automatic implantable cardioverter defibrillators: review of 12 years 25,000 patients. PACE Pacing Clin Electrophysiol 1993;16:174-8.[Medline]
   
4. Kelly PA, Cannom DS, Garan H, et al. The automatic implantable cardioverter defibrillator: efficacy, complications, and survival in patients with malignant ventricular arrhythmias. J Am Coll Cardiol 1988;11:1278-86.[Abstract]
   
5. Frame R, Brodman R, Furman S, et al. Long-term stability of defibrillation thresholds with intrapericardial defibrillator patches. PACE Pacing Clin Electrophysiol 1993;16:208-12.[Medline]
   
6. Lehman MH, Saksena S. NASPE policy statement: implanted cardioverter defibrillators in cardiovascular practice--report of the policy conference of the North American Society of Pacing and Electrophysiology. PACE Pacing Clin Electrophysiol 1991;24:969-79.
   
7. Herse B, Antschbach R, Hannekum A, et al. Different ways of application of the implantable cardioverter defibrillator (ICD): which is the current approach? Thorac Cardiovasc Surg 1992;40:266-72.
   
8. Goodman LR, Almass GH, Troup PJ, et al. Complications of automatic implantable cardioverter defibrillators: radiographic CT and echocardiographic evaluation. Radiology 1989;170:447-52.[Abstract/Free Full Text]
   
9. Grimm W, Flores BF, Marchlinski FE. Complications of implantable cardioverter defibrillator therapy: follow-up of 241 patients. PACE Pacing Clin Electrophysiol 1993;16:218-22.[Medline]
   
10. Mittleman RS, Mack K, Rastegar H, et al. Inappropriate shocks and elevation of defibrillation thresholds in a patient with automatic defibrillator patch Silastic erosion and titanium mesh fraying. PACE Pacing Clin Electrophysiol 1991;14:1452-5.[Medline]
    
11. Almassi GH, Olinger GN, Wetherbee JN, et al. Long-term complications of implantable cardioverter defibrillator lead systems. Ann Thorac Surg 1993;55:888-92.[Abstract]
    
12. Saksena S. Endocardial lead systems for implantable cardioverter defibrillators: uncertain progress beyond base camp. PACE Pacing Clin Electrophysiol 1992;15:123-5.[Medline]
    
13. Ideker RE, Wolf PD, Alferness C, et al. Current concepts for selecting the location, size, and shape of defibrillation electrodes. PACE Pacing Clin Electrophysiol 1991;14:227-40.[Medline]
    

This article has been cited by other articles:

				T.-Y. Hsia, S. M. Bradley, M. J. LaPage, S. Whelan, J. P. Saul, J. M. Ringewald, and J. H. Reed Novel Minimally Invasive, Intrapericardial Implantable Cardioverter Defibrillator Coil System: A Useful Approach to Arrhythmia Therapy in Children Ann. Thorac. Surg., April 1, 2009; 87(4): 1234 - 1239. [Abstract] [Full Text] [PDF]

				S. Suzuki, H. Watanabe, S. Yoshii, S. Kaga, Y. Honda, N. Ishikawa, and M. Matsumoto Alternative Technique for Implanting an Implantable Cardioverter Defibrillator in Infants Ann. Thorac. Surg., November 1, 2008; 86(5): 1701 - 1703. [Abstract] [Full Text] [PDF]

				H. Watanabe, J.-i. Hayashi, M. Haga, M. Saito, H. Suzuki, and S. Sato Successful implantation of a cardioverter defibrillator in an infant Ann. Thorac. Surg., December 1, 2001; 72(6): 2125 - 2127. [Abstract] [Full Text] [PDF]

				H. Mehmanesh, R. Bauernschmitt, and R. Lange Pectoral implantation of single-lead implantable cardioverter-defibrillators in patients without transvenous access Ann. Thorac. Surg., October 1, 1999; 68(4): 1426 - 1427. [Abstract] [Full Text] [PDF]

				U. Wolfhard and F. H. Splittgerber Intramural migration of implantable cardioverter-defibrillator patch J. Thorac. Cardiovasc. Surg., March 1, 1996; 111(3): 681 - 681. [Full Text]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): J. Ernesto Molina Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Molina, J. E. Articles by Adler, S. Search for Related Content PubMed PubMed Citation Articles by Molina, J. E. Articles by Adler, S.