Patients with a high defibrillation threshold: Clinical characteristics, management, and outcome

Of 125 patients prepared to receive implantable cardioverter-defibrillators (ICDs) with the patch-patch configuration of the defibriilating electrodes, 23 (18%) had high (225 joules) dgfibrlllation thresholds (DFTs). These patients had lower left ventricular ejection fractions (27 + 12 vs 34 + 13; p < 0.03) and a higher incidence of previous heart surgery (47% vs 19%; p < 0.01) than patients with normal DFTs but did not differ in age, type of heart disease, incidence of concomitant heart surgery, or use of antiarrhythmic medication. Defibrlllators were implanted in 18 of these 23 patients, 12 during the initial surgery and six after repeat DFT testing 2 weeks later. After 22 IT 11 months of follow-up, four patients with ICDs died (two suddenly, and two of nonsudden cardiac causes). Two patients without ICDs died of nonsudden cardiac causes. Appropriate shocks were received by five patients (29%) including both who died suddenly later. A high DFT may be more common than previously stated. It is associated with poor ventricular function and previous heart surgery. Repeated OFT testing may be useful in some patients. A high DFT does not preclude successful ICD shocks, but other therapies may provide better results. (AM HEART J 1991;122:89.)

Sergio L. Pinski, MD, Gabriel Vanerio, MD, Lon W. Castle, MD, Victor A. Morant, MD, Tony W. Simmons, MD, Richard G. Trohman, MD, Bruce L. Wilcoff, MD, and James D. Maloney, MD. Cleveland, Ohio

Implantable cardioverter-defibrillators (ICD) can generally prevent sudden death, even in high-risk patients.l However, sudden death in patients with ICDs has been documented by ECG in some in- stances to be the result of ineffective defibrillation.2 In a multicenter study ICD firing was observed in 77 % of witnessed sudden deaths, also suggesting in- effective defibrillation as the cause of those deaths.3

The probability of successful defibrillation is a sig- moidal-shaped function of energy rather a simple all-or-none (true “threshold”) phenomenon.4 During implantation of ICDs only the defibrillation thresh- old (DFT) (the energy associated with a high proba- bility of successful defibrillation)5-7 can be estimated. The goal is to find the greatest “margin of safety,” which is the difference between the DFT and the maximum output of the ICD (usually about 30 joules). A margin greater than 10 joules is considered safe. In some patients an acceptable DFT cannot be

From the Department of Cardiology, The Cleveland Clinic Foundation.

Received for publication July 17, 1990; accepted Dec. 20, 1990.

Reprint requests: James D. Maloney. MD, The Cleveland Clinic Founda- tion, Department of Cardiology. Desk F-15,9500 Euclid Ave, Cleveland, OH 4419.5.

411129353

achieved even after laborious operative maneuvers are used in positioning of the electrodes. Difficult de- cisions must then be made: Should the ICD never- theless be implanted, and what are the alternative therapeutic maneuvers? We analyzed the clinical characteristics and long-term outcome of 23 patients with DFTs equal to or greater than 25 joules mea- sured during attempted ICD implantation witli*,a patch-patch configuration of the defibrillating el&- trodes.

METHODS

Study population. Between 1984 and 1988, a total of 132 patients with malignant ventricular arrhythmias refrac- tory to antiarrhythmic drugs underwent surgery for im- plantation of an ICD with the patch-patch configuration of the defibrillating electrodes. Adequate intraoperative de- terminations of the DFT could be performed in 127 patients. This determination was not possible in two patients because hemodynamic instability precluded ex- tensive testing and in three because ventricular fibrillation could not be reliably induced. For the purpose of study ho- mogeny, two patients who received a Guardian ICD (Telec- tronics, Englewood, Colo.) were excluded from the analy- sis. The remaining 125 patients in whom implantation of an AICD (CPI, St. Paul, Minn.) was attempted constitute the study population.

89

90 Pin-ski et al. July 1991 American Heart Journal

surface and served initially as the anode. The other patch,

Operative procedures. Patients were placed under general anesthesia, which was usually maintained with ni-

sutured to the posterior or posterolateral left ventricle,

trous oxide and a narcotic. The lead system was placed through a left thoracotomy in 81 patients and through a

served initially as the cathode. In general, an attempt was

median sternotomy in 46. Of these 46 patients, 42 under- went concomitant heart surgery. In these 42 patients, the

made to place the patches intrapericardially. Two epicar-

ICD leads were installed and tested after other procedures were completed. In all instances patch electrodes were placed so as to include the largest mass of myocardium

dial sensing screw-in electrodes were also placed on the left

possible between them (e.g., points of contact between

ventricle.

patches were carefully avoided). Either small (14 cm2) or large (28 cm2) patches (CPI, St. Paul, Minn.) were used. The size and position of the initial electrodes were deter-

DFT testing. Ventricular fibrillation, defined as a ven-

mined by the operating surgeon. A number of factors influenced the decision regarding patch size including the location of aortocoronary vein grafts, internal mammary artery grafts, ventriculotomy, and overall heart size. When- ever possible at least one large patch was used.8 One patch (the small one when one small patch and one large patch were used) was sutured to the anterior right ventricular

comitant open-heart surgery, ICD implantation was de- ferred until the effects of surgery on the inducibiIity of ventricular tachyarrhythmia were assessed.g

Patients were considered to be on a regimen of amio- darone at the time of surgery when they had received the drug at a maintenance dose for at least 2 weeks (after a loading period). Results of determinations of plasma ami- odarone and desethylamiodarone levels (high-performance liquid chromatography) within 2 weeks of surgery (mean 2.5 days before) were available in 50 patients.

implanted. In four patients a spring lead-patch configuration was

tested at this time. A spring lead (Intec Systems, Pitts- burgh, Pa.) was inserted into the superior vena cava and paired to either of the epicardial patches. If a satisfactory

Management of patients with high DFTs. Management of patients with DFTs equal to or greater than 25 joules was

threshold was achieved the ICD was implanted in that

individualized. In general when the DFT was equal to 25 joules, the ICD was implanted. If the DFT was greater than

configuration. Decisions concerning further management

25 joules the ICD was not implanted, and the leads were capped and tunneled into the abdominal wound. Further DFT testing was done after several weeks, and if it was clinically feasible antiarrhythmic drugs were withdrawn or reduced in the interim. If at the time of the second test the DFT was still greater than 30 joules an ICD was not

tricular tachyarrhythmia during which QRS complexes could not be discerned in any of three simultaneously re- corded surface ECG leads, was induced by briefly applying alternating current to the heart. After 10 to 15 seconds de- fibrillation was attempted through an external cardiovert- er-defibrillator (Intec Systems, Pittsburgh, Pa.). The ini- tial energy tested was 15 joules. If that attempt was successful a 10 joule shock was then tried. Unsuccessful at- tempts were followed by a 30- to 40-joule “rescue shock.” If necessary further shocks were delivered through internal defibrillation paddles or transthoracically. When the initial l&joule shock was unsuccessful, increments of 5 joules were tested in subsequent trials until defibrillation was achieved. Defibrillation trials were separated from one an- other by at least 1 minute. The DFT was defined as the lowest energy that consistently terminated ventricular fi- brillation.

If the DFT with the initial patch configuration was not satisfactory (<20 joules), the entire test was repeated after repositioning the patches, inverting patch polarities, or both. In 13 patients in whom a satisfactory DFT was not achieved initially, the plasma potassium level was deter- mined intraoperatively and if necessary was brought to the 4 to 5 mmol/L range. If even then the desired DFT was not achieved, one or both patches were replaced with larger ones, space permitting. The total number of defibrillation trials was dictated by the patient’s hemodynamic stability.

When a satisfactory DFT was achieved and before the

of patients who did not receive ICDs were made by their personal physicians.

Postoperative defibrillation check. All patients except those in whom ICDs were not implanted at the same time as the patches were candidates for predischarge ICD func- tion testing. Ventricular fibrillation was induced in the electrophysiology laboratory by delivering alterating cur- rent through a transvenous catheter placed in the right ventricular apex. Effective defibrillation was defined as termination of ventricular fibrillation by the ICD with ei- ther the first or second shock.

Patient follow-up. Eighty-six patients were followed-up primarily by us. The remaining 36 patients were followed- up at other centers. Follow-up in these patients was by di- rect communication with the patient or with the patient’s physician. Patients were seen every 2 months for the first year after implantation and every month thereafter. At each visit symptoms, physical signs, and the number of shocks delivered by the ICD were recorded. A shock was considered appropriate when it was preceded by near-syn- cope, syncope, or ECG-documented ventricular tachyar- rhythmia. Patients received antiarrhythmic drugs as deemed clinically necessary.

Statistics. All data are presented as mean values k standard deviation. Comparisons were made by t test or chi-square test. Actuarial analysis of survival was per- formed according to Kaplan and Meier method. A p value of less than 0.05 was considered significant.

chest was closed, the ICD was tested at least once against induced ventricular fibrillation. In every patient the ICD RESULTS

recognized and terminated ventricular fibrillation with one lntraoperative DFT. Among 23 patients (18%) the or two shocks. In 21 of the 42 patients who underwent con- DFT was equal to or greater than 25 joules (25 joules

VOlUme 122

Number 1, Part 1 Management of high defibrillation threshold 91

Table I. Comparison between patients in group 1 (DFT 2 25 joules) and group 2 (DFT < 25 joules) -. - Group 1 In = 23) Group 2 (n = 1021 p Value

_- -.

Age (yr) 58 3~ 8 60 t 11

CAD (C;‘) 19 (82) 78 (77)

Sudden death ((‘c ) 12 (52) 41 (46)

Recurrent VT ( “1) 11 (48) 55 (54)

LVEF (? ) 27 + 12 34 + 13

Concomitant open-heart surgery ((r ) 6 (26) 36 (35) Previous open-heart surgery (% ) 11 (48) 19 (19) On antiarrhythmic agents (“A ) 18 (78) 86 (84) On Amiodarone ( !YE ) 13 (57) 62 (60) On class I drugs ( CU ) 11 (48) 56 (55)

On both amiodarone and class I (%) 6 (26) 31 (21) Amiodarone dose (mg) 462 f 150 448 IIZ 116 Amiodarone level (rig/ml) 1.9 + 1.1* 1.6 i 0.7t DEA level (@g/ml) 1.3 + 0.6* 1.1 i o.jt

CAD, coronary artery disease; DEA, desethylamiodarone; DFT, defibrillation threshold; LVEF, left ventricular ejection fractions.

NS N's NS NS

<0.03 NS 0.01 NS NS NS NS NS NS NS

*n = 9.

tn = 41.

in 10, 30 joules in five, and more than 30 joules in eight; group 1). The last patch configuration tested in these patients was large patch-large patch in seven (31% ) and large patch-small patch in 16 (69 % ). Both patches were inside the pericardium in 19 patients (83 % ). One patch was inside the pericardium and the other was outside in four patients (17 % ). The potas- sium level was within the normal range in the three patients in whom it was measured.

In the remaining 102 patients (82 % ) the DFT was less than 25 joules (group 2). Their final patch con- figuration was large patch-large patch in five (6 % ), large patch-small patch in 92 (90%), and small patch-small patch in four (4 % ). Both patches were inside the pericardium in 83 patients (81%). One patch was inside and the other was outside the peri- cardium in seven patients (7 % ). Both patches were extrapericardial in 12 patients (12 % ). The potassium level was measured in 10 patients. Two received in- traoperative potassium infusions with a subsequent decrease in the DFT. The incidence of a high DFT was 35% in 1986 (14 implants), 14% in 1987 (49 im- plants), and 19% in 1988 (59 implants). There was only one implant in each of the years 1984 and 1985.

Comparison between patients in group 1 and group 2. Patients in both groups did not differ significantly with regard to age, type of heart disease, clinical pre- sentation, incidence of concomitant surgery, surgical approach, or use of amiodarone, class I antiarrhyth- mic drugs, or both at time of implantation. Amio- darone dosage and plasma levels did not differ either (Table I). Patients in group 1 had a lower left ven- tricular ejection fraction (27 rt 12% vs 34 + 13% ; p < 0.03) and a higher incidence of previous heart surgery (47 % vs 19% ; p = 0.01). The ejection frac-

tion and the incidence of previous surgery did not differ significantly between patients operated on in 1986 and those operated on in 1987 and 1988. A DFT equal to or greater than 20 joules was found in 52 pa- tients (42 % ). These patients differed from those with DFTs less than 20 joules only in the incidence of pre- vious heart surgery (36% vs 15% ; p = 0.01).

Management of patients with high DFTs. An ICD was eventually implanted in 18 of the 23 patients in group 1 (Fig. 1). In 12 of them ICDs were implanted at the initial attempt (the DFT was 25 joules in 10 and 30 joules in two patients). In three patients no further attempts were made to implant an ICD. Two had a DFT of 40 joules and poor hemodynamic tolerance to defibrillation trials; repeated testing seemed unwar- ranted. The other patient went into recurrent ven- tricular fibrillation during intraoperative DFT test- ing and was resuscitated only with difficulty. Patches were removed because it was thought that they were contributing to electrical instability.

Eight patients underwent a second DFT test 15 + 3 days after the first attempt. Four were taking antiarrhythmic drugs. Six patients received an ICD at this time. In two patients with DFTs of 25 joules and 30 joules, a high-output ICD (36 joules) was im- planted. Four patients had extensive mediastinal and epicardial fibrosis (from previous cardiac surgeries) that was thought to be partly responsible for the high DFT. In these patients, after reconfirmation of a high DFT with the patch-patch configuration, a spring vena cava lead-patch configuration was tested. The DFT was lowered in three of them from 35 joules to 25 joules, from 40 joules to 25 joules, and from 30 joules to 15 joules. In the other patient the DFT was the same with both configurations (30 joules), and the

92 Pinski et al. July ,991

American Heart Journal

r ICD r DFT !25 J at initial teating

23 pts

not implanted 11 pts

1 No further testing

3 Rts Further testing

8 pts (DFT still

! 25 J in all)

No ICD 2 pts I

Spring lead tested 4 pte

Fig. 1. Flow chart showing management in 23 patients with high defibrillation threshold (DFT) at initial surgery. Implantable cardioverter-defibrillator @CD) was eventually placed in 18 patients.

Table II. Clinical characteristics and outcome of patients who did not receive defibrillators

Patient Age Heart L VEF Previous DFT Follow-up NO. (yr) disease (%) surgery (joules) Retested ho) Treatment Status

1 65 CAD 20 CABG 40 No 13 Amiodarone Dead: refractory VF (in-hospital)

2 59 CAD 38 CABG x2 40 No 17 Amiodarone Alive: no VT 3 54 VHD 28 AVR x2 40 No 15 Amiodarone Dead: heart

failure 4 55 PMD 11 None 40 Yes 15 Amiodarone Alive: awaiting

cardiac transplant 5 67 CAD 37 None 30 Yes 27 Quinidine Alive: no VT

AVR, aortic valve replacement; CABG, coronary artery bypass gafting; CAD, coronary artery disease; DFT, defibrillation threshold; LVEF, left ventricular ejection fraction; PMD, primary myocardial disease; VF, ventricular fibrillation; VHD, valvular heart disease; VT, ventricular tachycardia.

ICD was finally implanted in the patch-patch con- figuration. In two patients the DFT was still greater than 30 joules with the patch-patch configuration. No ICD was implanted.

Before the final decision to implant an ICD, group 1 patients underwent 8.1 & 4.8 defibrillation trials, whereas group 2 patients underwent a mean of 3.1 -t 1.7 trials (p < 0.000001). The clinical charac- teristics of patients who did not receive ICDs are given in Table II.

Results of postoperative defibrillation check. Predis- charge ICD testing was done in 11 of the 12 eligible patients in group 1 and in 72 of the 80 in group 2. The remaining nine candidates refused to undergo the procedure. Defibrillation was unsuccessful in one pa- tient in group 1 (25 joules, ICD implanted at the pri- mary attempt) and in three patients in group 2. In these four patients the chest was reopened, and after

further repositioning of the patches the ICD effec- tively terminated induced ventricular fibrillation.

Follow-up. Patients were followed up for 25 + 11 months. There were six deaths in group 1. Sixteen (89 % ) of the 18 patients with ICDs in this group were maintained on a regimen of antiarrhythmic drugs (amiodarone alone in 11, class I drugs alone in four, and a combination of amiodarone and class I drugs in five). There were four deaths among them; two were nonsudden cardiac deaths and two were sudden un- monitored deaths. Relatives of one of these patients witnessed four ICD discharges after the patient col- lapsed. Five of the 18 patients (29%) received ap- propriate ICD shocks, including those two who died suddenly later.

Patients who did not receive ICDs were treated medically. Two of them died, one of heart failure and the other of refractory ventricular fibrillation in the

volume 122

Number 1, Part 1 Management of high defihrillatic~,r threshold 93

intensive care unit 1 day after combined peripheral revascularization and foot-amputation surgery. One of t,he three surviving patients is in end-stage heart failure and is awaiting a cardiac transplant. Seventy- four (72 % ) patients in group 2 were maintained on a regimen of antiarrhythmic drugs (amiodarone alone in 39, class I drugs alone in 22, and a combination of amiodarone and class I drugs in 13). There were 18 deaths in this group. Five were noncardiac, 11 were cardiac nonsudden, and two were sudden. Appropri- ate shocks were received by 38 patients (37 % ).

Actuarial survival at 1, 2, and 3 years was 82 % , 74%,and74% ingroupland91%,86% and78% in group 2, respectively (not significant). Actuarial incidence of sudden death at 2 years was 12% in pa- tients in group 1 with ICDs and 2% in patients in group 2 @ < 0.05). Incidence of appropriate shocks did not differ significantly between patients in group 1 with ICDs and patients in group 2.

DISCUSSION

The DFT is commonly used as a clinical estimate of the ability to defibrillate.4-7 Common sense and clinical datalo support aiming for a DFT equal to or less than 15 joules during ICD implantation. Proto- cols used to determine the DFT at different institu- tions vary widely.7 It is not known whether any of them is more clinically relevant, but standardization would facilitate comparing results between institu- tions.

High DFTs were thought to be uncommon with the use of the patch-patch electrode configuration. It has been stated that in approximately 5 % of patients no configuration reliably converts ventricular fibrilla- tion.6 Our incidence of high DFT (18 % ) with a patch-patch configuration seems rather high. The highest incidence (35 % ) during 1986 may represent our “learning curve,” but the incidence of approxi- mately 15% during 1987 and 1988 may be accurate for our current techniques. Certainly every effort was made by the surgical team to achieve an optimum DFT, as reflected by the large number of defibrilla- tion trials in these patients. A likely explanation for our high incidence of high DFTs may be an overall higher prevalence of variables associated with high DFTs in our patients.

We found that patients in group 1 had a signifi- cantly lower left ventricular ejection fraction and a significantly higher incidence of previous heart sur- gery. Depressed left ventricular function has been correlated with higher DFTs in somel but not all8 studies. The relationship between myocardial mass and DFT also remains controversial.ll, l2 We cannot speculate on which variable is a better predictor of

high DFT, since we did not routinely measure mYo-

cardial mass or chamber size. Preoperatively severely impaired ventricular function, gross cardiomegaly, or both should make the physician aware of the possi- bility of finding a high DFT.“’

The relationship between previous cardiac surgery and high DFTs has not been reported before. Inter- estingly, when a cutoff point of 20 joules was used, the presence of previous heart surgery remained the only variable differing significantly between patients with DFT above or below that value. We hypothesize that the epicardial and mediastinal scarring in these pa- tients interferes with correct positioning of patches, acts as insulation during epicardial defibrillation, or both.

The effects of pharmacologic interventions on car- diac defibrillation have been the subject of recent re- search, but because of the nonuniformity of design among experimental studies and the fragmentary nature of clinical reports, no firm conclusions can be drawn at this time. l4 We found no association be- tween use of class I antiarrhythmic agents, amio- darone, or both and a high DFT. In previous studies lidocaine and class IC drugs consistently increased the DFT, whereas procainamide and quinidine had varied effects.15 In our study few patients were tak- ing each of the different class I agents, precluding meaningful comparisons. The association between amiodarone and high DFTs suggested by other investigators8. 16, l: cannot be ruled out definitely from our study. It is still possible that in individual patients with marginal DFTs, amiodarone may cause an unexpectedly marked increase in the defibrilla- tion energy requirements. Until more information is collected it may be wise when possible to retest de- fibrillation in the absence of antiarrhythmic drugs in patients with high DFTs.

Simple operative maneuvers such as repositioning of patches, exchanging small patches for larger ones,6 or reversing patch polarityls2 I9 are useful for decreas- ing a high DFT. The best course of action when they fail is unclear. If another type of therapy is not fea- sible, the generator may nevertheless be implanted in the hope that the DFT will decrease with the passage of time.20 Alternatively, implantation can be deferred until a lower DFT is documented. during repeat testing.6 Both strategies were used in this series. We implanted ICDs in most patients with a DFT of 25 joules. In other patients we deferred implantation, but this strategy was not especially rewarding, inas- much as the DFT did not decrease spontaneously in any patient. The costs of washing out antiarrhythmic drugs in the hospital were substantial. Unexpectedly, a lower DFT was achieved with a spring lead-patch

94 Pin-ski et al.

configuration in three of four patients in whom scar- ring from previous cardiac surgery made intraopera- tive positioning the patches difficult.

Overall mortality did not differ significantly be- tween group 1 and group 2. However, the incidence of sudden death in patients in group 1 with ICDs was significantly higher than that in patients in group 2. Five patients in group 1 received potentially life-sav- ing shocks. But the two instances of sudden death, both in patients with previous appropriate shocks, attest to the partial protection given by the ICD in patients with high DFTs. Recent data from Lehmann et a1.3 also suggest a disproportionate incidence of high DFTs in patients with ICDs who died suddenly. Cardiac transplantation may be a therapeutic alter- native for patients with high DFTs, particularly those with severe ventricular dysfunction. The sub- optimum protection provided by the ICD could be used as a “bridge” to transplantation.21

Limitations of the study. The retrospective uncon- trolled nature of our study represents its main limi- tation. The probable incidence of our “learning curve” in the results has already been stated. Never- theless, we believe that the study faithfully reflects clinical practice. “Normal” DFT values have not been determined. Our definition of a high DFT (25 joules or more) is an arbitrary one, which is based on our goal of achieving a safety margin of at least 10 joules during implantation. When 20 joules was used as the cutoff point, only the presence of previous heart surgery remained as a predictor of a high DFT. Our DFT testing protocol did not allow us to stratify patients into other DFT ranges.

Future developments. Several modifications that augment the efficiency of internal defibrillation will be available in the next generation of ICDS.~~~ 23 With these sophisticated devices high DFTs should be- come less of a problem. On the other hand, with the expected proliferation of ICDs (eventually boosted by the availability of a nonthoracotomy system24), many ICDs may be implanted by physicians with less than ideal expertise. This may result in an increase in the frequency of complications. Sound clinical judgement, familiarity with the ICDs being used, and close attention to the details of DFT testing will con- tinue to be necessary for the best patient care.

Conclusions. High DFTs, even with the patch- patch configuration, may be more common than pre- viously stated. High DFTs are associated with de- pressed left ventricular function and previous cardiac surgery. Deferring implantation after repeated test- ing (including a spring lead configuration in selected patients) may be useful. Protection against sudden death given by the ICD is suboptimum, so other

July 1991 American Heart Journal

forms of therapy may be indicated in some of these patients.

We thank Mr. Tom Lang for his assistance in the preparation of this manuscript.

REFERENCES

1.

2.

Troup PJ. Implantable cardioverter and defibrillators. Curr Probl Cardiol 1989;14:675-815.

3.

4.

5.

6.

7.

a.

9.

10.

Luceri RM, Habal SM, Castellanos A, Thurer RJ, Waters RS, Brownstein SL. Mechanism of death in patients with the au- tomatic implantable cardioverter defibrillator. PACE 1988; 11:2015-22. Lehmann MH, Thomas A, Jackson K, et al. Long-term outcome with implantable cardioverter-defibrillator (ICD) therapy in a multicenter investigator edited database‘ [Ab: stractl. Circulation 1990;82:111-166. Davy JM, Fain ES, Do& P, Winkle RA. The relationship between successful defibrillation and delivered energy in open-chest dogs: reappraisal of the “defibrillation threshold” conceut. AM HEART J 1987:113:77-83. Rattes MF, Jones DL, Sharma AD, Klein GJ. Defibrillation threshold: a simple and quantitative estimate of the ability to defibrillate. PACE 1987;10:70-7. Cannom DS, Winkle RA. Implantation of the automatic im- plantable cardioverter defibrillator (AICD): practical aspects. PACE 1986;9:793-809. Lehmann MH, Steinman RT, Schuger CD. Jackson K. De- fibrillation threshold testing and other practices related to AICD implantation: do all roads lead to Rome? PACE 1989;12:1530-7. Troup PJ, Chapman PD, Olinger GN, Kleinman LH. The im- planted defibrillator: relation of defibrillating lead configura- tion and clinical variables to defibrillation threshold. J Am Co11 Cardiol1985;6:1315-21. Manolis AS, Rastegar H, Estes III NAM. Prophylactic auto- matic implantable cardioverter defibrillator patches in pa- tients at high-risk for postoperative ventricular arrhythmias. J Am Co11 Cardiol 1989;13:1367-73. Marchlinski FE, Flores BF, Miller BF, Gottlieb CD, Hargrove 1111 WC. Relation of intraoperative defibrillation threshold to successful postoperative defibrillation with an automatic im- plantable cardioverter-defibrillator. Am J Cardiol 198862: 393-8.

11. O’Donoghue S, Platia EV, Mispireta L, Goldstein S, Wacloski S. Relationships between left ventricular mass and ejection fraction, and defibrillation threshold [Abstractl. J Am Co11

12.

13.

14.

15.

16.

17.

18.

Cardiol.l990;15:51A. Chapman PD, Sagar KB, Wetherbee JN, Troup PJ. Relation- ship of left ventricular mass to defibrillation threshold for the implantable defibrillator: a combined clinical and animal study. AM HEART J 1987;114:274-8. Vlay SC. Defibrillation testing: necessary but evil? AM HEART J 1989;117:499-504. Ruffy R. Pharmacological modulation of ventricular defihril- lation. In: Zipes DP, Jalife J, eds. Cardiac electrophysiology: From cell to bedside. Philadelphia: WB Saunders Company, 1990:959-62. Singer I, Guarnieri T, Kupersmith J. Implanted automatic defibrillators: effects of drugs and pacemakers. PACE 1988; 11:2250-62. Guarnieri T, Levine JH, Veltri EP, et al. Success of chronic defibrillation and the role of antiarrhythmic drugs with the automatic implantable cardioverter-defibrillator. Am J Car- diol 1987;60:1061-4. Frame LH. The effect of chronic oral and acute intravenous amiodarone on ventricular defibrillation threshold using im- planted electrodes in dogs. PACE 1989;12:339-46. O’Neill P, Lawrie G, Harvill L, Pacific0 A. Significance of

Volume 122

Number 1, Part 1

19.

20.

21.

22.

patch polarity on defibrillation threshold [Abstract]. J Am Co11 Cardiol 1990;15351A. Troup PJ, Wetherbee JN, Chapman PD, et al. Does electrode polarity affect defibrillation efficacy [Abstract]? PACE 1990;13:528. Fain ES, Billingham M, Winkle R. Internal cardiac defibrilla- tion: hi&pathology and temporal stability of defibrillation energy requirements. J Am Co11 Cardiol 1987;9:631-8. Bolling SF, Deeb GM, Morady F, et al. AICD: a new “bridge” to cardiac transplantation [Abstract]. J Am Co11 Cardiol 1990;15:223A. Saksena S. Implantable antitachycardia devices: the next

23.

24.

Management of high, defibrillation threah,old 95

generation. In: Saksena S, Goldschlager h. eds. Electrical therapy for cardiac arrhythmias. Philadelphia: WB Saunders Company, 1990:554-73. Winkle RA. The implantable defibrillator: progression from first- to third-generation devices. In: Zipes DP, Jalife J, eds. Cardiac electrophysiology: from cell to bedside. Philadelphia: WB Saunders Company, 1990:963-9. Saksena S, Tullo NG, Kool RB, Mauro AM. Initial clinical ex- perience with endocardial defibrillation using an implantable cardioverter-defibrillator with a triple electrode system. Arch Intern Med 1989;149:2333-9.