Comparison of efficacy of automatic implantable cardioverter defibrillator in patients older and younger than 65 years of age

Comparison of Efficacy of Automatic Implantable Cardioverter Defibrillator in Patients Older and Younger Than 65 Years of Age DONALDD. TRESCH, M.D., PAUL J. TROUP, M.D., RAN JAN K. THAKUR, M.D., JAN VESETH-ROGERS, R.N., VlCKIE TUCKER, R.N., JULE N. WETHERBEE, M.D., RAYMOND G. HOFFMAN, Ph.D., PETER D. CHAPMAN, M.D., Milwaukee, Wisconsin

PURPOSE: The efficacy of the automatic implantable cardioverter defibrillator (AICD) was compared in elderly patients and younger patients with life-threatenlng ventricula~- tachyarrhythmig~. Clinical characteristics, surgical complications, and long-term survival rates were compared between the two age groups. PATIENTS AND METHODS: A retrospective study was conducted of 54 elderly patients (greater than 65 years) and 79 younger patients (less than 65 years) who had had AICDs implanb~d for rec u r r e n t symptomatic ventHcui~r tachycardia a n d / o r ventric-lnr fibrillation. RESULTS: In 85% of elderly patients and 78% of younger patients, coronary a r t e r y disease was the underlying disease (NS). The mean left ventricnlar ejection fraction was 31.4 + 14.3% in the elderly patients and 35.7 ± 17.6% in the younger patients (NS). Concomitant myocardial revascularization was performed in 37 % of elderly patients and 29% of younger patients (NS); however, only 4% of elderly patients had concomitant left ventHcul~r resection or cryoablation, compared with 15% of younger patients (p <0.001). Two patients in each age group died peHoperatively (4% versus 3%, NS), and no significant difference in surgical morbidity or length of hospital stay following AICD implantation was noted between the age groups. In conjunction with AICD, elderly patients more commonly received antiarrhythmic drugs, with 54% of elderly patients t~ldng amiodarone at the time of hospital dischRrge compared with 29% of the younger patients (p <0.008). In contrast, ~blockers were more commonly used in younger From the Departmentsof Cardiology(DDT, PJT, RKT,JVR, VT, JNW), Geriatrics/Gerontology(DDT), and Biostatistics(RGH), MedicalCollege of Wisconsin, Milwaukee,Wisconsin. Requestsfor reprints should be addressedto DonaldD. Tresch, M.D., Divisionof Cardiology,MedicalCollegeof Wisconsin, 8700 West Wisconsin Avenue,Milwaukee,Wisconsin 53226. Dr. Troup's current address: Mt. Sinai Medical Center, Milwaukee, Wisconsin. Manuscript submitted April 26, 1990, and accepted in revised form March 13, 1991.

patients (16% versus 2%, p <0.03). At a mean follow-up of 25 months, 11 (20%) elderly patients and 16 (20%) younger patients had died. Six deaths in elderly patients and five deaths in younger patients were classified as arrhythmlc deaths (NS); however, only one younger patient and three elderly patients died suddenly (NS). Calculated survival curves demonstrated similar survival rates in the two age groups with approYimately 90%, 87%, and 80% of the patients alive at 1, 2, and 3 years, respectively. Theoretic survival curves calculated from appropriate AICD shocks demonstrated significantly lower survival compared with actual survival. CONCLUSION: It is concluded that AICD is a very effective treatment for life-threatening ventricular t a c h y a r r h y t h m i ~ , and this benefit applies to elderly patients as well as younger patients.

he automatic implantable cardioverter defi-

T brillator (AICD) is now accepted therapy for recurrent symptomatic ventricular tachycardia and/or ventricular fibrillation. During the last 5 years, the use of the AICD has rapidly increased with more than 10,000 devices being implanted [1,2]. Results with the device have been impressive, with multiple studies demonstrating survival from sudden cardiac death to be 98% at I year and 94% at 5 years compared with previous 1-year survival rates of 65% in patients with such lethal arrhythmias [3-8]. These impressive results have led some authorities to suggest the use of the AICD as the first-line approach in patients with malignant ventricular arrhythmias [9]. Application of the AICD, however, is complex and requires preoperative electrophysiologic evaluation, heart surgery, and frequent follow-up evaluations. Besides initial implantation, generators require replacement due to battery depletion, and leads may require revision [6,10]. Whether such therapy is justified in elderly patients, who, in general, are known to have higher surgical mortality, June 1991 The American Journal of Medicine Volume 90

717,

CARDIOVERTER DEFIBRILLATORS IN THE ELDERLY/ TRESCH ET AL

require longer hospital stays, and are more prone to complications from aggressive therapies, has not previously been investigated. In an attempt to determine the efficacy of AICD in elderly patients compared with younger patients with recurrent symptomatic ventricular tachycardia and/or ventricular fibrillation, we undertook the current study. Patients receiving AICDs were divided into two age groups: elderly (greater than 65 years) and younger (less than 65 years). Clinical characteristics, surgical complications, including length of hospitalization, and long-term survival rates were compared between the two age groups.

PATIENTS AND METHODS A total of 133 consecutive patients who had AICDs implanted from June 13, 1983, through March 8, 1988, for recurrent symptomatic ventricular tachycardia and/or ventricular fibrillation were retrospectively studied. All but one patient had their initial AICD inserted at a Medical College of Wisconsin-affiliated institution. Patients were divided into two age groups: a younger group, which was comprised of patients less than 65 years old (n = 79) and an elderly group, which was comprised of patients greater than 65 years old (n = 54). Patients were selected for insertion of AICD if they had sustained either cardiac arrest or symptomatic ventricular tachycardia and if they demonstrated by electrophysiologic testing inducible ventricular tachycardia or ventricular fibrillation that could not be suppressed by antiarrhythmic drug therapy. The AICD was also inserted in patients who had sustained cardiac arrest, although ventricular tachycardia or ventricular fibrillation could not be induced by electrophysiologic testing. The selection criteria were the same for both age groups. Prior to AICD insertion, all patients underwent cardiac catheterization with coronary angiography and electrophysiologic evaluation. Presence of coronary artery disease was defined as at least one obstructive lesion of 70% or greater reduction in luminal diameter in one of the three major coronary arteries. Left ventricular ejection fractions were calculated from the ventriculogram or from radionuclide ventriculography. Electrophysiologic testing was performed at one or more right ventricular sites with introduction of one, two, or three extra stimuli at three drive cycles. Inducible ventricular fibrillation was defined as a ventricular tachyarrhythmia with absence of clearly defined QRS complexes on the body surface electrocardiogram. Inducible sustained ventricular tachycardia was defined as an organized ventricular arrhythmia with a cycle length greater than 200 milliseconds that lasted 30 seconds or more or required cardioversion. Inducible nonsustained ven718

June 1991 The American Journal of Medicine

Volume 90

tricular tachycardia was defined as ventricular tachycardia (monomorphic or polymorphic) that lasted 30 seconds or less. A patient's arrhythmia was considered noninducible if only five or fewer ventricular re-entrant beats could be induced. The surgical technique for AICD implantation was similar to that previously reported [11]. A left anterior thoracotomy approach through the fifth or sixth interspace was used unless concomitant cardiac surgery requiring cardiopulmonary bypass was performed. Conventional defibrillation electrode systems were used, and defibrillation thresholds were determined at the time of surgery before the AICD device was attached to the electrode leads. Concomitant surgery consisted of coronary artery bypass grafting, left ventricular endocardial resection, and/or left ventricular cryoablation. Implantation of the AICD was performed at the time of the concomitant cardiac surgery or was performed in a staged manner. The staged approach consisted of placement of AICD electrodes at the time of a cardiac surgical procedure designed to treat" the arrhythmia primarily and confirmation that defibrillation energy requirements were less than or equal to 20 J. Following the initial surgical procedure, an AICD pulse generator was implanted if the patient exhibited spontaneous ventricular tachycardia or if drugresistant inducible ventricular a r r h y t h m i a rem a i n e d a t t h e t i m e of t h e p o s t o p e r a t i v e electrophysiologic testing. Prior to insertion of the AICD, patients who were capable of exercise underwent maximal exercise treadmill testing to establish the maximum heart rate and select the AICD cutoff rate. Antiarrhythmic drug therapy in combination with AICD implantation was determined prior to AICD generator implantation: drugs were used to suppress the spontaneous occurrence of atrial and/or ventricular arrhythmias, to prolong ventricular tachycardia cycle length, to make its induction more difficult, or to reduce the rate of the ventricular response in patients with supraventricular tachyarrhythmias. After discharge from the hospital, patients were followed at 1- or 2-month intervals in the cardiac arrhythmia clinic by one of the authors. At each clinic visit, the AICD device was interrogated to determine the number of delivered shocks. Patients who experienced AICD discharges were interviewed regarding the circumstances of the shock, and associated symptoms such as palpitations, syncope, or near syncope were determined. Shocks were considered appropriate (due to recurrent malignant ventricular arrhythmia) if hypotensive ventricular tachycardia or ventricular fibrillation was documented by electrocardiography at the time of the shock and/or if the patient experienced syncope. Shocks associated with electrocardiographic docu-

CARDIOVERTER DEFIBRILLATORS IN THE ELDERLY / TRESCH ET AL

TABLEI Clinical Characteristicsof Patients Age Groups

ClinicalCharacteristics

Younger(n = 79)

Elderly (n = 54)

p Value

Ventricular fibrillation Recurrentventricular tachycardia Underlying cardiac disease Coronary artery disease Single vessel Two-vessel Three-vessel LV aneurysm Primary cardiomyopathy Primary electrical disease ProlongedQ-T syndrome Valvular heart disease Mean LVEF0.30 or <0.30 Maximum heart rate (beats/minute)t

58 (73%) 21 (27%)

31 (57%) 23 (43%)

NS NS

61 (78%) ] 0 (16%) ] 6 (26%) 35 (58%) 33 (42%) 12 (14%) 3 (4%) 1 (1%) 2 (3%) 35 (44%) 139 -+ 20

46 (85%) 5 (10%) 9 (20%) 32 (70%) 16 (30%) 8 ( ] 5%) 0 0 1 (2%)* 30 (56%) 130 t 15

NS NS NS NS NS NS NS NS NS NS <0.05

LV = leftventricle;LVEF= leftventricularejectionfraction. *Aorticvalvulardiseaseplusprimarycardiomyopathy. 'At timeof stresstesting.

mented rhythms other than ventricular tachycardia or fibrillation were considered inappropriate and all other shocks were classified as unknown. Deaths were classified according to previous repbrted studies [3] of AICD use: (1) sudden if death was known to be within 1 hour of onset of symptoms or if it was unwitnessed; (2) arrhythmic if it was due to a documented tachyarrhythmia but was not sudden or unwitnessed; (3) cardiac death if it was due to a cardiac etiology but was not sudden or arrhythmic; and (4) noncardiac if it was due to a noncardiac etiology. Total arrhythmic deaths included sudden and nonsudden arrhythmic deaths. Statistical Analysis All data are expressed as the mean ± 1 SD. Comparisons were made using the unpaired t-test or Fisher's exact test where appropriate. Both actuarial and theoretic survival curves were calculated by the life-table method, using the AICD generator implant date as the starting date. Theoretic survival of a patient was based on the time interval from implantation of the device to the occurrence of the first appropriate shock or until death. Survival curves were estimated by the Kaplan-Meier method. The relative risk was estimated using the proportional hazards model; and the 95% confidence interval for the relative risk was estimated using the standard errors from the proportional hazards model [12]. A relative risk was considered significant at the 0.05 level if the 95% confidence interval did not include 1.0.

and the mean age of the younger group was 52.8 410.4 years. Twenty-five elderly patients (46%) were older than 70 years and eight (15%) were older than 75 years. Nineteen younger patients (24%) were younger than 50 years and eight (10%) were younger than 40 years. Twenty (37%) elderly patients were women, compared with 35 (44%) of the younger patients (NS).

Pre-AICD Insertion Table I shows the clinical characteristics and the specific underlying cardiac disease in each age group. Younger patients more commonly experienced ventricular fibrillation prior to AICD insertion, although the difference did not reach statistical significance. The majority of patients in each age group exhibited coronary artery disease with abnormal left ventricular function. Systolic ventricular function was poorer in the elderly, compared with the younger group, although the difference did not reach statistical significance (left ventricular ejection fractions: 31.4 4- 14.3% versus 35.7 ± 17.6%). Fifty-six percent of elderly patients and 44% of younger patients had left ventricular ejection fractions of 30% or less (NS). Electrophysiologic testing results were similar in both groups with 31% of elderly patients and 27% of younger patients classified as having noninducible arrhythmias at the time of baseline testing (NS). At the time of exercise stress testing prior to AICD implantation, younger patients exhibited significantly higher maximum heart rates (Table I).

f

RESULTS Demographics The mean age of the entire group of 133 patients was 59.7 4- 11.9 years (range: 14 to 80 years). The mean age of the elderly group was 69.8 4- 4.3 years,

AICD Insertion Ninety (68%) of the 133 patients had the AICD implanted as an isolated surgical procedure. Twenty elderly patients (37%) and 23 younger patients (29%) had the AICD implanted in conjunction with June 1991

The American Journal of Medicine Volume 90

719

CARDIOVERTER DEFIBRILLATORS IN THE ELDERLY / TRESCH ET AL

TABLE II AICDImplantation AgeGroup Younger(n = 79)

Elderly (n = 54)

p Value

56 (71%) 20 (25%) 12 (15%)

34 (63%) 20 (37%) 2 (4%)

NS NS <0.001

34 (43%) 8 (10%) 6 (8%) 7 (9%) 4 (5%) 1 (1%) 23 (29%) lO (16%) 12.2 _+ 6.2

9 (17%) 8 (15%) 7 (13%) 3 (6%) 9 (17%) 6 (11%) 29 (54%) 1 (2%) 11.4 _+ 5.8

< 0.003 NS NS NS < 0.04 <0.02 < 0.008 <0.03 NS

11.7 --8.9 18.7 -+ 14.4

10.1 _+ 7.2 20 _+ 16.2

NS NS

Surgery AICD alone CABG LV resection +/or cryoablation Conjunctive antiarrhythmic drug therapy No drugs Quinidine Procainamide Flecainide Mexiletine Tocainide Amiodarone B-Blockers Defribillator threshold (joules) Length of hospital stay (days)* AICD alone Concomitant procedure CABG= coronaryarterybypassgrafting;LV : leftventricle. *Hospitalstayfollowingdeviceimplantation.

another cardiac surgical procedure (Table II). No significant difference was noted in the use of coronary revascularization between the age groups; however, left ventricular endocardial resection was performed in 10 younger patients (in combination with coronary revascularization in eight patients), whereas only two elderly patients had a left ventricu|~r endocardial resection (both in combination with coronary revascul~rization). Left ventriclllar cryoablation was performed in five younger patients; none of the elderly patients had cryoablation performed. Defibrillation thresholds were slrni|ar in both age groups (Table 1I).

Use of antiarrhythmic drug therapy in conjunction with AICD implantation differed significantly between the age groups (Table II). Eighty-three percent of elderly patients continued to take an antiarrhythmic drug (not including fl-blockers or calcium channel blockers) at the time of hospital discharge, compared with only 57% of the younger patients (p <0.003). Amiodarone was used in 29 elderly patients (54%) and 23 younger patients (29%) (p <0.008). Sixteen percent of younger patients were taking fl-blockers at the time of hospital discharge, compared with 2% of elderly patients (p
TABLE III

Post-AICD Insertion

Complications of Long-Term Follow-Up*

HOSPITALCOURSE:Four patients (two elderly and two younger) died during hospitalization following implantation of the AICD. One elderly patient, a 74-year-old man, died suddenly i day after surgery of ventricular fibrillation that could not be converted by either the AICD or an external defibrillator; the other elderly patient, a 73-year-old man, died of recurrent refractory ventricular tachycardia 2 days postoperatively. The two deaths in the younger patients were due to pulmonary emboli, 9 and 10 days after surgery, respectively. In both of these patients, the AICD was inserted in conjunction with myocardial revascularization and/or left ventricular resection. Aside from the four deaths, other perioperative complications were infrequent in both age groups (Table III). Three elderly patients sustained cerebral vascular accidents postoperatively; however, only one of the three patients had permanent sequalae. Heart failure occurred in two elderly

Younger in = 79) Perioperativecomplications Total deaths Sudden Arrhythmic (non-sudden) Noncardiacdeaths Cerebralvascular accident Post-pericardotomysyndrome Adult respiratory distress syndrome Drug-induced hepatitis Late complications Total deaths Sudden Arrhythmic (nonsudden) Nonarrhythmic cardiac Noncardiac Infections Constrictive pericarditis Appropriate shocks Inappropriateshocks Deviceexplanted or deactivated

Elderly (n = 54)

2 0 0 2 0 2 1 1

2 1 1 0 3 0 0 0

14 1 4 6 3 2 0 15 36 7

9 2 2 2 3 1 1 19 16 4

*Noneof thedifferences between theagegroupswassignificant.

720

June 1991

The American Journal of Medicine Volume 90

CARDIOVERTER DEFIBRILLATORS IN THE ELDERLY / TRESCH ET AL

patients and in one younger patient. The adult respiratory distress syndrome, which occurred in one younger patient, was thought to be related to amiodarone toxicity, and the patient required temporary ventilatory support. Length of hospital stay following AICD implantation was not significantly different between the age groups (Table II). As expected, length of hospital stay was significantly longer in both age groups when the AICD was inserted in conjunction with another cardiac surgical procedure; however, even in these cases, the length of hospital stay was similar between the age groups.

N

79

67

61

52

39

28

23

18

13

7

3

Sudden I

0.9

Damtha

Nonsudden Arrhythmlc i Nonarrhythmlc Cardiac

L

0.8

L

L

Noncardlac

0.7 S U 0,6

I Theoretical Deaths

vR 0.5

l__

~ 0.4 0.3

Younger Patients

0.2

}

0.1 0.0 0

6

12

18

54

42

37

29

Follow-Up Table III lists the late complications of the two age groups. Eleven patients (four elderly and seven younger) had the AICD explanted or deactivated. Four of the seven younger patients underwent cardiac transplantation; one younger patient refused replacement of the AICD; in another younger patient, the AICD was deactivated because the patient was terminally ill due to lung cancer; and in one younger patient with refractory heart failure, the device was deactivated when it became clear that a transplant donor could not be found. In the four elderly patients, two AICD generators were not replaced because the patients became severely incapacitated due to progressive cerebral disease; in another patient, the AICD was removed at the time of a pericardectomy for constrictive pericarditis, which was thought to be due to exuberant fibrosis surrounding the patch electrodes [13]; and in one patient, the device was deactivated because of repeated inappropriate firing of the AICD due to interaction with a dual chamber permanent pacemaker. Two younger patients and one older patient developed an infection of the AICD pocket, and in each case the AICD had to be removed and replaced by a new device. In one patient, the infection occurred within the first month after AICD insertion, and in the other two patients, the infection occurred within 7 months after insertion of the device. At a mean follow-up of 25.1:1= 17.2 months (range: 1 to 66 months), there have been 23 deaths in addition to the four surgical deaths. Sixteen younger patients and 11 elderly patients have died (Table III). One younger patient and three elderly patients died suddenly; three of these sudden deaths were unwitnessed and one occurred postoperatively. All of the sudden deaths occurred within 1 year after implantation of the AICD. In addition to the four patients who died suddenly, four other younger patients and three other elderly patients died from

46

1.0

N

24 30 36 42 48 MONTHS POST IMPLANT

23

12

10

6

3

54

60

66

2

1

1

72

1.0

Sudden Deaths Nonmudden Arrhythmlc

0.90.8-

I

I

0.7-

L_

S U 0.6R

Nonlrrhythmlc

L

Cardiac

N.....

d,,c

7

V 0.54 v A

0.4~

L 0.3

[ Elderly Patient8 ]

0.2

Theoretical Deaths

[ _ _

0.1 0.0 o

6

12

18

24 30 36 42 48 MONTHS POST IMPLANT

54

60

66

72

Figure 1. Comparison of actual and theoretic survival curves between the age groups. Survival curves are classified accordin 8 to specific cause of death.

arrhythmias. The AICD was documented to function properly at the time of death in six of these seven patients; in one elderly patient with recurrent ventricular tachycardia, the AICD did not fire. No significant difference was noted in the total atrhythmic deaths (sudden and nonsudden) between the two age groups. Other cardiac complications were the cause of death in eight patients: three younger patients and two elderly patients died of refractory heart failure, and three younger patients died of cardiogenic shock, endocarditis, and left ventricular rupture, respectively. Five younger patients and three elderly patients died of noncardiac causes. During the follow-up period, seven younger patients were designated cardiac transplantation candidates. Four of these seven patients underwent transplantation; one patient died of heart failure before receiving a transplanted heart; and two are presently still awaiting transplantation.

June 1991 The American Journal of Medicine Volume 90

721

CARDIOVERTER DEFIBRILLATORS IN THE ELDERLY / TRESCH ET AL

TABLEIV

Predictorsof Mortality* Total Morality Clinical Variable Age < 65 years > 65 years Gender Male Female CAD Present Absent Amiodarone yes No LVEF < 30% > 30% Concomitant surgery Yes No Noninducible arrhythmia at baseline EPS Yes No

Total Arrhythmic Death'

RR

95% Cl

RR

95% CI

1.0 2.1

1.01-4.5'

1.0 3.2

0.9-12

1.8 1.0

0.5-4.8

1.9 1.0

0.2-16

1.3 1.0

0.2-2.6

4.7 1.0

0.4-54

1.4 1.0

0.6-3.3

1.0 1.0

0.4-2.2

2.2 1.0

1.3-5.3'

2.4 1.0

0.9-6.4

1.0 1.4

0.7-2.5

1.0 1.4

0.4-5.0

1.0 2.5

0.8-10

1.0 2.0

0.7-5.0

CAD = coronaryarterydisease;CI = confidenceinterval;EPS = electrophysiologicstudy; LMEF= lef~ventdcularejectionfraction;RR = relativerisk. * Probabilityof survivalestimatedfrom Kaplan-Meiersurvivalcurves.Numberof suddendeathsin eachgroupwas toofew to be statisticallyanalyzed. 'Includessuddenand non-suddenarrhythmicdeaths. ' Statisticallysignificant.

Sixty-five percent of both age groups received AICD shocks after hospital discharge following AICD implantation, and no significant difference was noted between the age groups with regard to the number of appropriate shocks received (Table III). In analyzing appropriate shocks in reference to the effect of concomitant heart surgery, no significant difference was noted in the number of appropriate shocks received between patients with and without concomitant heart surgery. Figure I shows the actual and theoretic survival curves for each age group. In each age group, a significant difference was noted between theoretic and actual survivals (elderly, p <0.007; younger, p <0.04). At 3 years' follow-up, only one younger patient and three elderly patients have died suddenly and approximately 80% of patients in each group are alive. Clinical variables were examined for all 133 patients to assess their ability to predict outcome independently (Table IV). Left ventricular function was found to be the best predictor of total mortality, with an absolute risk of 0.72 probability of survival at 24 months in patients whose left ventricular ejection fraction was 30% or less compared with a 0.92 probability of survival in patients whose ejection fraction was greater than 30%. The estimated relative risk was 2.2 (CI 1.3 to 5.3, p <0.02). Patient's age was also a predictor of total mortality, although it was marginal and less significant than

722

June 1991

The American Journal of Medicine

Volume 90

the left ventricular ejection fraction. The estimated relative risk was 2.1 (CI 1.01 to 4.5, p <0.049). None of the clinical variables, including patient's age or left ventricular function, was predictive of sudden death or total arrhythmic death. COMMENTS

The results of this study are similar to the results reported by Tresch et al [14] in an earlier study in which benefits and complications of aggressive evaluation and therapy were compared between elderly patients and younger patients with recurrent symptomatic ventricular tachyarrhythmias. In that study, there was no significant difference in complications from cardiac catheterization and anglographic studies or from repeated electrophysiologic studies between patients older than 65 years comparedwith patients younger than 55 years. A higher cardiac surgical mortality was noted in the older patients, although long-term survival was similar in both age groups. In the current study, which included only patients receiving AICDs, regardless of other therapy, there was no significant difference in surgical mortality and/or morbidity between the age groups. The postoperative length of hospital stay following AICD implantation was not significantly different between the groups. Interestingly, previous studies of coronary artery bypass surgery [15,16] have reported length of hospital stay to be

CARDIOVERTERDEFIBRILLATORSIN THE ELDERLY/ TRESCH ET AL

dependent on the patient's age, with elderly patients requiring longer stays. In this study, even when AICD implantation accompanied another cardiac surgical procedure, postoperative length of stay was similar for both age groups. Long-term survival and the percentage of patients receiving appropriate AICD shocks after AICD implantation were also similar between the age groups. A difference in the type of concomitant cardiac surgery in conjunction with AICD insertion was noted between the age groups. Younger patients more often had left ventricular subendocardial resection or cryoablation prior to AICD implantation. This undoubtedly reflects the bias of the investigators who were influenced by the prior experience of higher perioperative mortality experienced by elderly patients, along with the fact that elderly patients had worse left ventricular systolic function, which also influenced decision-making regarding attempts at curative arrhythmia surgery. Another difference noted in the treatment of the two age groups was in the use of antiarrhythmic drug therapy following the insertion of the AICD. Elderly patients more often received maintainance therapy with antiarrhythmic drugs, with approximately 50% of the elderly patients discharged with a regimen of amiodarone. Regarding the difference in amiodarone use, again, investigator bias was a likely factor. Intolerance of antiarrhythmic drugs, along with poor left ventricular systolic function, often limited the therapeutic options in the elderly population. Younger patients generally tend to be more tolerant of drug therapy aside from amiodarone and, since amiodarone toxicity is time dependent [17], it may be reasonably assumed that younger patients will be exposed to the drug for longer periods. On the other hand, amiodarone toxicity has been shown to be associated with advanced age [18]. ~-Blockers were much more commonly used in younger patients, largely because of the better systolic left ventricular function in that group. In addition, maximum sinus rate has been shown to be agerelated [19,20] and, as would be expected, younger patients demonstrated a significantly faster maximum heart rate during stress testing in this study. Therefore, it was more often necessary to use/3blockers in the younger patients to prevent inappropriate AICD firing due to overlap between sinus tachycardia and ventricular tachycardia. The overall results of our study are comparable to previous reports [3-8] of other centers' experiences with AICD in the treatment of recurrent ventricular tachyarrhythmias and support the findings that the AICD is very effective in preventing sudden death. The 2-year survival rate of approximately 90% in

both age groups in our study is higher than that reported with the use of antiarrhythmic drug therapy alone [21,22] or in combination with surgery [23,24]. And the 3-year survival rate of approximately 80% is even more encouraging. At the 3-year follow-up, only one of the younger patients and three of the elderly patients had died suddenly. Furthermore, in comparing these results with theoretic survival curves, we noted a significant improvement in the actual survival. As has been reported in previous studies [25,26], left ventricular function was the best predictor of total mortality; patient's age was also predictive of total mortality, although marginal and less significant than left ventricular function. None of the clinical variables, including patient's age or left ventricular function, was an independent predictor of sudden death or total arrhythmic death. Interestingly, four patients in our study underwent cardiac transplantation after a mean followup of 18 months following AICD insertion, and three other patients were scheduled for cardiac transplantation during the study. Even though all patients in our study had documented symptomatic ventricular tachyarrhythmias, use of the AICD as a "bridge" to cardiac transplantation has been recently reported [27]. It is known that many deaths in patients awaiting cardiac transplantation are sudden [28,29], and AICD insertion may protect certain patients with asymptomatic ventricular tac h y a r r h y t h m i a s w h o are a w a i t i n g c a r d i a c transplantation. The limitations of the study need to be emphasized. The study was retrospective, and, even though specific patient selection criteria were used, patient selection was partially subjective. Only elderly patients who were functionally active prior to the development of their ventricular tachyarrhythmia would have undergone the aggressive evaluation prior to making a decision concerning the use of an AICD. Therefore, the elderly patients in our study would have to be considered a select group of elderly patients with life-threatening arrhythmias. Our patient selection in reference to the use of concomitant heart surgery and the use of amiodarone was also subjective, in that we were reluctant to subject elderly patients to left ventricular resection, although we more readily used amiodarone in these patients. It should also be noted that there were certain differences in outcome between the age groups: three elderly patients sustained cerebral vascular accidents postoperatively, whereas none of the younger patients had any neurologic complications following surgery. Furthermore, two elderly patients had the AICD explanted due to progressive

June 1991

The American Journal of Medicine Volume 90

723

CARDIOVERTER DEFIBRILLATORSIN THE ELDERLY / TRESCH ET AL

neurologic impairments. Nevertheless, in general, elderly patients who were selected to have an AICD implanted due to life-threatening ventricular tachyarrhythmias had an excellent prognosis. We conclude that AICD is very effective in the treatment of life-threatening ventricular tachyarrhythmia and this benefit applies to elderly patients as well as younger patients. A difference in the type of concomitant surgery and in the use of antiarrhythmic drugs is noted between the age groups; however, most elderly patients are capable of tolerating the aggressive evaluation and surgery, and improved long-term survival is as impressive in elderly patients as in younger patients.

REFERENCES 1. Troup PJ. Implantable cardioverters and defibrillators. Current Problems in Cardioloogy 1989: 14: 679-815. 2. Manolis AS, Rastegar H, Estes NA. Automatic implantable cardioverter defibrillator. JAMA 1989; 262: 1362-8. 3. Winkle RA, Mead RH, Ruder MA, etaL Long-term outcome with the automatic implantable cardioverter defibrillator. J Am Coil Cardiol 1989; 13: 1353-61. 4.Tchou PJ, Kadri N, Anderson J, et aL Automatic implantable cardioverter defibrillators and survival of patients with left ventricular dysfunction and malignant ventricular arrhythmias. Ann Intern Med 1988; 109: 529-34. 5. Kelly PA, Cannom DS, Garan H, et aL The automatic implantable cardioverterdefibrillator: efficacy, complications and survival in patients with malignant ventricular arrhythmias. J Am Coil Cardiol 1988; 11: 1278-86. 6. Thomas AC, Moser SA, Smutka ML, Wilson PA. Implantable defibrillation: eight years clinical experience. PACE 1988; 11: 2053-8. 7. Mirowski M, Mower MM, Staewen WS, Tabatznik B, Mendeloff AI. Standby automatic defibrillator: an approach to prevention of sudden coronary death. Arch Intern Med 1970; 126: 158-61. 8. Platia EV, Griffith LSC, Watkins L, et al. Treatment of malignant ventricular arrhythmias with endocardial resection and implantation of the automatic cardioverter-defibrillator. N Engl J Med 1986; 314: 213-6. 9. Lehmann MH, Steinman RT, Schuger CD, Jackson K. The automatic implantable cardioverter defibrillator as antiarrhythmic treatment morbidity of choice for survivors of cardiac arrest unrelated to acute myocardial infarction. Am J Cardiol 1988; 62: 805-13. 10. Kadri N, Niazi I, Elkhatib I, et aL Automatic implantable cardioverter defibrillator: problems and complications [abstract]. J Am Coil Cardio11987; 9: 142A. 11. Troup PJ, Chapman PD, Olinger GN, Kleinman LH. The implanted defibrillator: relation of defibrillating lead configuration and clinical variables to defibrillation threshold. J Am Coil Cardiol 1985; 6: 1315-21.

724

June 1991 The American Journal of Medicine Volume 90

12. Kleinbaum DS, Kupper LL, Morgenstern H. Epidemiology research: principles and quantitative methods. Belmont, California: Lifetime Learning Publications, 1982. 13. Almassi GH, Chapman PD, Troup PJ, Wetherbee JN, Olinger GN. Constrictive pericarditis associated with patch electrodes of the automatric implantable cardioverter-defibrillator. Chest 1987; 92: 369-71. 14. Tresch DD, Platia EV, Guarnieri T, et aL Refractory symptomatic ventricular tachycardia and ventricular fibrillation in elderly patients. Am J Med 1987; 83: 399-404. 15. RoseDM, GelbfishJ, Jacobowitz IS, e t aL Analysis of morbidity and mortality in patients 70 years of age and over undergoing isolated coronary artery bypass surgery. Am Heart J 1985; 110: 341-6. 16. Gersh BJ, Kronmal RA, Frye RL, et aL Coronary arteriography and coronary artery bypass surgery: morbidity and mortality in patients ages 65 years or older. Circulation 1983; 67: 483-90. 17. Smith WM, Lubbe WF, Whitlock RM, et al. Long-term tolerance of amiodarone treatment for cardiac arrhythmias. Am J Cardiol 1986; 57: 1288-93. 18. Herre JM, Sauve MJ, Malone P, et al. Long-term results of amiodarone therapy in patients with recurrent sustained ventricular tachycardia or ventricular fibrillation. J Am Coil Cardiol 1989; 13: 442-9. 19. Rodeheffer RJ, Gerstenblith G, Becker L, et aL Exercise cardiac output is maintained in healthy human subjects: cardiac dilution and increased stroke volume compensated for in diminished heart rate. Circulation 1984; 69: 203-12. 20. Morley JE, ReesSS. Clinical implications of the aging heart. Am J Med 1989; 86: 77-86. 21. Schaffer WA, Cobb LA. Recurrent ventricular fibrillation and modes of death in survivors of out-of-hospital ventricular fibrillation. N Engl J Med 1975; 293: 259-62. 22. Lampert S, Lown B, Grayboys TB, Podrid PJ, Blatt C. Determinants of survival in patients with malignant ventricular arrhythmias associated with coronary artery disease. Am J Cardiol 1988; 61: 791-7. 23. Swerdlow CD, MasonJW, Stinson EB, et al. Results of operations for ventricular tachycardia in 105 patients. J Thorac Cardiovasc Surg 1986; 92: 105-13. 24. McGiffin DC, Kirklin JK, Plumb VJ, et aL Relief of life-threatening ventricular tachycardia and survival after direct operations. Circulation 1987; 76: V93-103. 25. Freedman RA, Swerdlow CD, Soderholm-Difatte V, Mason JW. Prognostic significance of arrhythmia inducibility or noninducibility at initial electrophysiologic study in survivors of cardiac arrest. Am J Cardiol 1988; 61: 578-82. 26. Poole JE, Mathisen TL Kudenchuk PJ, et aL Long-term outcome in patients who survive out of hospital ventricular fibrillation and undergo electrophysiologic studies: evaluation of electrophysiologic subgroups. J Am Coil Cardiol 1990; 16: 657-65. 27. Boiling SF, Deeb GM, Morady F, et aL AICD: a new "bridge" to cardiac transplantation [abstract]. J Am Coll Cardiol 1990; 15: 223A. 28. Stevenson LW, Chelimsky-Fallick C, Tillisch J, et aL Unacceptable risk of sudden death without transplantation if low ejection fraction is due to coronary artery disease [abstract}. J Am Coil Cardiol 1990; 15: 222A. 29. Keogh AM, Baron DW, Hickie JB. Prognostic guides in patients with idiopathic or ischemic dilated cardiomyopathy assessedfor cardiac transplantation. Am J Cardiol 1990; 65: 903-9.