lACC Vol. 7, No 4 Apnl 1986:925-32

CARDIAC PACING

925

Comparative Survival After Permanent Ventricular and Dual Chamber Pacing for Patients With Chronic High Degree Atrioventricular Block With and Without Preexistent Congestive Heart Failure

MARTIN A, ALPERT, MD, FACC, JACK J. CURTIS, MD, FACC, JOHN F. SANFELIPPO, MD,

GREG C. FLAKER, MD, FACC, JOSEPH T. WALLS, MD, FACC, VASKAR MUKERJI, MD,

DANIEL VILLARREAL, MD, S. K. KATTI, PHD, NIALL P. MADIGAN, MD, FACC,

RYSZARD B. KROL, MD

Columbia, Missouri

To determine whether survival after permanent ven•tricular demand (VVI) pacing differs from survival after permanent dual chamber (DVI or DDD) pacing in pa•tients with chronic high degree atrioventricular (AV) block (Mobitz type II or trifascicular block), 132 patients who received a VVI pacemaker (Group 1) and 48 pa•tients who received a DVI or DDD pacemaker (Group 2) were followed up for 1 to 5 years. There was no significant difference in sex distribution, mean age or incidence of coronary heart disease, hypertension, val•vular heart disease, diabetes mellitus, stroke or renal failure between Groups 1 and 2.

Overall, the predicted cumulative survival rate at 1, 3 and 5 years was 89, 76 and 73%, respectively, for Group 1 and 95, 82 and 70%, respectively, for Group 2. In patients with preexistent congestive heart failure,

Before permanent cardiac pacemakers became available, patients with chronic Mobitz type II and complete trifas•cicular atrioventricular (A V) block had a dismal prognosis. Nearly 50% died within 1 year and 75 to 90% died within 5 years of the onset of these conduction disturbances (1-3). Those who lived were prone to frequent episodes of syn•cope, presyncope, congestive heart failure and a variety of symptoms attributable to low cardiac output (3). Many stud-

From the Departments of Medicine and Surgery, University of Missouri Health Sciences Center and Harry S. Truman Memorial Veterans Hospital, Columbia, Missouri.

Manuscript received August 29, 1985; revised manuscript received November 6, 1985, accepted November 19, 1985.

Address for reprints: Martin A. Alpert, MD, Division of Cardiology, Room lE-65, University of Missouri Health Sciences Center, One Hospital Drive, Columbia, Missouri 65212

© 1986 by the Amencan College of CardIology

the predicted cumulative survival rate at 1, 3 and 5 years was 85,66 and 47%, respectively, for Group 1 (n = 53) and 94, 81 and 69%, respectively, for Group 2 (n = 20). The 5 year predicted cumulative survival rate was significantly lower in Group 1 patients with preexistent congestive heart failure than in Group 2 patients with the same condition (p < 0.02). There was no significant difference in 5 year cumulative survival rate between Groups 1 and 2 for patients without preexistent conges•tive heart failure.

The results suggest that permanent dual chamber pacing enhances survival to a greater extent than does permanent ventricular demand pacing in patients with high degree A V block and preexistent congestive heart failure.

(J Am Coli CardioI1986;7:925-32)

ies (4-15) have demonstrated that permanent ventricular pacing has not only improved the quality of life of such individuals, but has also markedly improved survival, par•ticularly in those with preexistent congestive heart failure. Dual chamber pacing theoretically offers comparative hemodynamic advantages over ventricular pacing because it preserves atrial transport function, thus maintaining car•diac output at or near prebradyarrhythmic levels. Preser•vation of atrial transport function may particularly benefit individuals with left ventricular dysfunction. Whether the hemodynamic advantages of dual chamber pacing translate into improved survival is uncertain. The objective of this study was to compare survival of patients with and without congestive heart failure who received a permanent ventric•ular or dual chamber pacemaker for chronic Mobitz type II or trifascicular block.

0735-10971861$350

926 ALPERT ET AL SURVIVAL AFfER PERMANENT PACEMAKER IMPLANTATION

Methods Patient selection. Patients were considered eligible for

the study if they received a permanent cardiac pacemaker (VVI, DVI, DDD) for chronic persistent or intermittent high degree AV block. High degree AV block was defined as Mobitz type II A V block or trifascicular block (third degree A V block characterized by wide QRS complexes and a ventricular rate of less than 40 beats/min). Patients with acute or reversible precipitating causes, or both, for high degree A V block (including drugs, acute myocardial in•farction and cardiac surgery) and patients who were not pacemaker-dependent after implantation were excluded from the study. Information was obtained by interview with and examination of patients, review of the medical records and consultation with referring physicians.

Identification and classification of chronic coexistent illnesses. Congestive heart failure was defined in accord•ance with the criteria for definite heart failure used by McKee et al. (16) in the Framingham study. Patients were included only if congestive heart failure had preceded the heart block (preexistent congestive heart failure). Criteria for coronary heart disease included definite angina pectoris or a history of myocardial infarction confirmed by serum cardiac en•zyme elevation or characteristic electrocardiographic ab•normalities. Systemic hypertension was diagnosed when blood pressure exceeded 140/90 mm Hg on three consecutive out•patient visits. Congenital heart disease and dilated cardio•myopathy were confirmed by cardiac catheterization. Val-

Table 1. Patient Characteristics

All Patients

JACC Vol 7. No 4 Apnl 1986925-32

vular heart disease was diagnosed by echocardiography or cardiac catheterization. Patients with no evidence of organic heart disease were classified as having conduction system disease only. Diabetes mellitus was diagnosed on the basis of an elevated fasting blood sugar level.

Patient characteristics. The study group consisted of 180 patients; 132 with a permanent ventricular demand pace•maker (VVI) (Group 1) and 48 with a permanent dual cham•ber pacemaker (36 with DVI and 12 with DDD) (Group 2). Patient characteristics were analyzed for the entire study group and for the subgroups with preexistent congestive heart failure. Sex distribution, mean age and presenting arrhythmias are shown in Table 1. The nature and distri•bution of chronic coexistent illnesses are summarized in Table 2. Table 3 lists the therapeutic modalities used to treat the major chronic coexistent illnesses.

Follow-up. The follow-up period for Group 1 began on January 1,1966 and ended on December 31,1984; that for Group 2 began on January 4, 1979 and ended on December 31, 1984. Total follow-up time ranged from 1 to 18 years for Group 1 and from 1 to 5 years for Group 2. Group 1 and Group 2 patients who were lost to follow-up remained in the study until the time of their last clinic visit. Follow•up information was obtained by review of medical records, patient interviews and examination, discussions with fam•ilies and consultation with referring physicians.

Pacemaker information. Pulse generators and pace•maker electrodes in Group 1 patients were manufactured by Medtronic Inc. in 106 cases, Cordis Inc. in 14 cases and

Patients With CHF

Ventricular Dual Chamber Ventricular Dual Chamber Patient Characteristics Paced (Group I) Paced (Group 2) Paced (Group I) Paced (Group 2)

No. of patients 132 48 53 20 Men 75 32 31 12 Women 57 16 22 8

Mean age (yr) 67 ± 6 69 ± 6 68 ± 5 68 ± 6 Presenting arrhythmia

Mobitz type II A V block 35 II 9 5 Trifascicular block 118 46 49 17 Both* 21 9 5 2

Presenting symptoms Lassitude/weakness 93 35 39 13 Dyspnea on exertion 71 25 44 14 Pre-syncope 48 16 18 6 Syncope 45 16 19 9 Altered sensorium 20 9 10 4 Palpitation 9 2 3 I Angina pectoris 5 2 Seizures 3 0 0 0 No symptoms 14 5 0 0

*Refers to patients who demonstrated both trifascicular and Mobitz type II A V block at various times during the monitoring period. A V = atrioventricular; CHF = preexistent congestive heart failure.

lACC Vol. 7, No.4 Apnl 1986'925-32

Table 2. Chronic Coexistent Illnesses

ALPERT ET AL. 927 SURVIVAL AFTER PERMANENT PACEMAKER IMPLANTATION

All Patients Patients With CHF

Ventncular Coexistent Illness Paced (n = 132)

Conduction system 49(37)

disease only Congestive heart failure 53(40)

Coronary artery disease 40(30)

Diabetes mellitus 32(24)

Hypertension 17(13)

Valvular heart disease 12(9)*

Cerebral infarctlon 7(5)

Cancer 6(4)

Dilated cardiomyopathy 4(3)

Emphysema 4(3)

Renal failure 3(2)

ArteriosclerosIs obliterans 3(2)

Hypothyroidism 3(2):1:

Atrial septal defect 2(2)

Syphilis I( I)

ParkinSOnism I( I)

Tuberculosis I( I)

Dual Chamber Paced (n = 48)

18(38)

20(42)

20(42)

12(25) 8(16) 5(10)t

2(4) 1(2)

1(2) 3(6) 0(0)

2(4) 1(2):1: 0(0) 0(0)

0(0) 0(0)

Ventricular Paced (n = 53)

28(53)

\0(19)

8(15)

7(13) 1(2)

1(2) 4(8) 1(2)

3(6) 2(4) 1(2)

0(0) 0(0)

0(0)

0(0)

Dual Chamber Paced (n = 20)

9(43)

6(30) 3(15)

1(5) 1(5)

1(5) 1(5) 1(5) 0(0)

1(5) 0(0)

0(0)

0(0) 0(0)

0(0)

* Seven with aortic stenosis, four with mitral regurgitation, one with mitral stenosis and two with previous valve replacement; t four with aortic stenosis, one with mitral regurgitation; :j: euthyroid when studied. Numbers in parentheses signify percent of group or subgroup. CHF = congestive heart failure.

Starr-Edwards Inc, in 12 cases. Pulse generators and elec•trodes in Group 2 patients were manufactured by Medtronic Inc. in all cases.

Statistical analysis. Survival curves were derived using methods described by Gehan (17). Breslow's generalized Kruskal-Wallis test (18) was used to determine whether there were differences in predicted cumulative survival be•tween subgroups. Survival data were analyzed for the first 5 years of follow-up in Groups 1 and 2 to permit valid comparison of data. The chi-square test was used to deter•mine whether differences in the incidence of presenting arrhythmias, presenting symptoms or chronic coexistent ill•nesses between Groups 1 and 2 were significant. A prob•ability (p) value of less than 0.05 was considered necessary for statistical significance.

Results Overall survival. Predicted cumulative survival rates

for years 1 through 5 for Groups 1 and 2 are shown in Table 4 and Figure 1. There was no significant difference in overall 5 year predicted cumulative survival rates between the two groups.

Survival in patients with and without preexistent congestive heart failure. Predicted cumulative survival rates for years 1 through 5 for patients in Groups 1 and 2 with preexistent congestive heart failure are shown in Table 4. Survival curves for patients in Groups 1 and 2 with and

without preexistent congestive heart failure are shown in Figures 2 and 3.

In Group 1, the 5 year predicted cumulative survival rate was significantly lower in patients with than in patients without preexistent congestive heart failure (p < 0.005)(Fig. 2). There was no significant difference in the 5 year pre•dicted cumulative survival rates between patients with and without congestive heart failure in Group 2 (Fig. 3). The predicted cumulative survival rate at 5 years was signifi•cantly higher in Group 2 patients with preexistent congestive heart failure than in Group 1 patients with preexistent failure (p < 0.02) (Fig. 4). There was no significant difference in predicted cumulative survival rates for patients without pre•existent congestive heart failure between Groups 1 and 2.

Causes of congestive heart failure. Causes of conges•tive heart failure in Group 1 patients were coronary artery disease in 28, hypertension in 11, valvular heart disease in 7, dilated cardiomyopathy in 4 and renal failure in 3 patients. Causes of congestive heart failure in Group 2 patients were coronary artery disease in 12, hypertension in 4, valvular heart disease in 3 and dilated cardiomyopathy in 1 patient.

Impact of other chronic coexistent illnesses on sur•vival. Predicted cumulative survival rates for years 1 through 5 for patients with coronary artery disease, hypertension, valvular heart disease and diabetes mellitus in Groups 1 and 2 are shown in Table 4. Other chronic coexistent illnesses occurred with insufficient frequency (Table 2) to calculate meaningful survival data. There were no significant differ-

928 ALPERT ET AL SURVIVAL AFfER PERMANENT PACEMAKER IMPLANTATION

JACC Vol 7, No 4 April 1986:925-32

Table 3. Therapeutic Modalities Used to Treat Major Chronic Coexistent Illnesses

Ventricular Paced Dual Chamber Paced

All Patients Patients With CHF All Patients Patients With CHF Therapy (n= 132) (n=53) (n=48) (n=20)

Medical Nitrate 39(30) 26(49) 18(38) 9(45) Beta-adrenergic 31(23) 0(0) 14(29) 0(0)

blocking drug Calcium channel 4(3) 2(4) 5(10) 4(20)

antagonist Digitalis 51(39) 51(100) 20(42) 20(100) Diuretic 52(39) 48(90) 23(48) 20(100) Arteriolar dilator 5(4) 2(4) 6(12) 4(20) Sympatholytic drug 5(4) 2(4) 4(8) 2(10) Insulin 22(17) 4(8) 8(\7) 2(10) Oral hypoglycemic drug 8(6) 2(4) 2(4) 1(5) Antineoplastic drug 6(4) 0(0) 1(2) 0(0) Adrenal corticosteroid 2(2) 0(0) 0(0) 0(0) Bronchodilator 4(3) 1(2) 3(6) 1(5) Sympathomimetic drug 2(2) 1(2) 1(2) 0(0) Thyroid supplement 3(2) 1(2) 0(0) 1(5)

Surgical/mechanical Coronary artery 5(4) 1(2) 3(6) 0(0)

bypass surgery Heart valve 2(2) 2(4) 1(2) 1(5)

replacement Hemodialysis 2(2) 2(4) 0(0) 0(0) Radiation therapy 1(1) 0(0) 0(0) 0(0)

There were no significant differences between ventricular (Group I) and dual chamber (Group 2) paced patients for any therapeutic modality either for all patients or for patients with congestive heart failure (CHF). Numbers in parentheses signify percent of group or subgroup.

ences in predicted cumulative 5 year survival rates between Groups 1 and 2 for patients with and without coronary artery disease, hypertension, valvular heart disease and diabetes mellitus. Of the 40 Group I patients with coronary artery disease, 13 had hypertension and 10 had diabetes mellitus. Of the 20 Group 2 patients with coronary artery disease, 5 had hypertension and 4 had diabetes mellitus.

Group 1 patients with isolated conduction system disease demonstrated significantly higher 5 year predicted cumu•lative survival rates than those in Group 1 with associated coronary artery disease (p < 0.01), hypertension (p < 0.025) or diabetes mellitus (p < 0.025) (Table 4). All Group 2 patients with isolated conduction system disease survived the follow-up period, thus precluding calculation of pre•dicted cumulative survival and preventing valid comparison with other subgroups using the Kruskal-Wallis test (Table 4).

Mortality and causes of death. Mortality data and cause of death in Groups I and 2 are summarized in Table 5.

Impact of sex and age at implantation on sur•vival. Table 4 shows the predicted cumulative survival rates for years 1 through 5 for Groups I and 2, for men and women and for patients less than 66 years old, 66 to 74

years old and 75 years old or more at the time of pacemaker implantation. There were no significant differences in pre•dicted cumulative survival between men and women in either group. Survival rates tended to decrease with increasing age, but the differences between groups were not statistically significant.

Discussion Prognosis after permanent ventricular pacing. The

prognosis of patients with untreated high degree A V block is grim. Reported survival rates at I year approach 50% and range from 10 to 33% at 5 years (1-3). Substantially higher survival rates have been reported after permanent ventricular pacemaker implantation, ranging from 81 to 93% at I year and 48 to 66% at 5 years (4-15). Although there is some variation in reported survival rates, virtually all studies as•sessing the prognosis of patients with high degree A V block after permanent ventricular pacemaker implantation have demonstrated that survival is closely related to the nature and severity of the underlying organic heart disease, pri•marily coronary artery disease and its complications (4-15).

JACC Vol 7. No.4 April 1986'925-32

Table 4. Survival Data

Mean Age Variable n (yr)

All patients Ventricular 132 67 ± 6 Dual chamber 48 69 ± 6

Men Ventricular 75 67 ± 6 Dual chamber 32 69 ± 6

Women Ventricular 57 68 ± 7 Dual chamber 16 69 ± 7

< 66 years old Ventricular 35 Dual chamber II

66 to 74 years old Ventricular 43 Dual chamber 16

2:: 75 years old Ventricular 54 Dual chamber 21

Mobitz type II AV block

Ventricular 35 67 ± 6 Dual chamber II 69 ± 6

Trifascicular block VentrIcular 118 67 ± 6 Dual chamber 49 69 ± 6

Conduction system disease only

Ventricular 49 67 ± 6 Dual chamber 18 69 ± 5

Congestive heart failure

VentrIcular 53 67 ± 5 Dual chamber 20 68 ± 6

Coronary artery disease

Ventricular 40 68 ± 6 Dual chamber 20 69 ± 6

Hypertension Ventricular 17 68 ± 7 Dual chamber 8 66 ± 9

Va\ vular heart disease

Ventricular 12 65 ± 10 Dual chamber 5 68 ± 8

Diabetes mellitus Ventricular 32 67 ± 6 Dual chamber 12 69 ± 8

AV = atrioventricular; n = number of patients.

Impact of preexistent congestive heart failure on sur•vival after permanent ventricular pacing. Several recent studies (13,15) have reported results that underscore the negative impact of preexistent congestive heart failure on survival after permanent ventricular pacemaker implanta•tion. Simon and Janz (15) reported a survival rate of 87, 68 and 55%, respectively, at 1, 3 and 5 years for patients

ALPERT ET AL. 929 SURVIVAL AFfER PERMANENT PACEMAKER IMPLANTATION

Predicted Cumulative Survival (%)

I yr 2 yr 3 yr 4 yr 5 yr

89 82 76 69 63 95 88 82 76 70

89 82 76 69 63 95 88 82 75 69

90 83 77 70 64 94 89 83 78 72

90 84 78 71 65 98 92 86 80 74

87 80 74 68 62 96 91 85 79 74

92 84 77 69 61 92 85 79 72 65

88 81 75 69 63 95 88 82 76 70

89 92 76 69 63 95 88 82 76 70

93 90 86 83 80

85 76 66 56 47 94 87 81 74 69

88 78 69 59 50 90 83 76 69 62

89 82 76 69 63 92 87 81 76 70

80 76 72 68 64

96 88 80 72 65 92 86 80 74 68

with high degree A V block and preexistent congestive heart failure, and a survival rate of 89, 77 and 66%, respectively, at 1, 3 and 5 years for patients without preexistent congestive heart failure. These differences were not statistically sig•nificant. Laczkovics et al. (19) reported a 50% mortality rate at 3 years and Castberg (20) reported a 50% mortality rate at 4 years for patients with congestive heart failure who

930 ALPERT ET AL SURVIVAL AFfER PERMANENT PACEMAKER IMPLANTATION

-:.e ~ 100 Q) .. nI a:: iii 80 > .; ... :::I 60 en Q) > ::: nI 40 :i E :::I (.)

20 1) ...... Ventricular-paced, n = 132 Q) .. 0--0 Dual chamber-paced, n = 48 u =s Q) 0 ...

Q. 2 3 4 5

Years Following Pacemaker Implantation

Figure 1. Overall predicted cumulative survival. There was no significant difference in overall 5 year predicted cumulative sur•vival rates between Group I (ventricular paced) and Group 2 (dual chamber paced).

received a permanent pacemaker for high degree A V block. In our study, the predicted cumulative survival rate at 5 years for patients with preexistent congestive heart failure (47%) was significantly lower than that of patients without preexistent failure (73%, p < 0.005). Thus, our results confirm the negative impact of preexistent congestive heart failure on survival after permanent ventricular pacemaker implantation for high degree A V block.

Hemodynamic advantages of dual chamber pacing over ventricular pacing. Dual chamber pacing has been shown to possess comparative hemodynamic advantages over

Figure 2. Predicted cumulative survival rates of ventricular paced patients (Group 1) with and without preexistent congestive heart failure (CHF). The 5 year predicted cumulative survival rate was significantly higher in patients without preexistent congestive heart failure (non-CHF) than in those with preexistent congestive heart failure (p < 0.005).

~ 100 Q) .... nI a:: 80 iii > 'e :::I 60 en Q) > ::: as 40 :i E :::I

...... CHF; n=53 (.) 20 1) 0--0 Non-CHF; n = 79 Q) .... p < 0.005 u =s

0 Q) ... 1 2 3 4 5 Q.

Years Following Pacemaker Implantation

~ 100 Q) .. as a:: 80 iii > .; ... :::I 60 en Q) > ::: as 40 :i E :::I (.) 20 1) Q) .. u =s

0 Q) .. Q.

...... CHF; n=2O

0--0 Non-CHF; n = 28

2 3 4

JACC Vol 7. No 4 Apnl 1986925-.12

5

Years Following Pacemaker Implantation

Figure 3. Predicted cumulative survival rates of dual chamber paced patients (Group 2) with and without congestive heart failure. There was no significant difference in the 5 year predicted cu•mulative survival rates between groups with and without preex•istent congestive heart failure (CHF) in patients receiving a per•manent dual chamber pacemaker.

ventricular pacing during acute myocardial infarction (21) and in patients recovering from cardiac surgery (22). Recent studies (23-30) have demonstrated that patients receiving a permanent dual chamber pacemaker may derive similar hemodynamic benefits. which accrue from properly timed atrial contraction. With ventricular pacing. atrial contraction enhances ventricular filling intermittently (with A V disso•ciation) or not at all (with 1:1 ventriculoatrial conduction). The resultant loss of atrial transport function is particularly

Figure 4. Predicted cumulative survival rates of patients with preexistent congestive heart failure who received either a perma•nent ventricular or dual chamber pacemaker. The 5 year predicted cumulative survival rate of patients with a permanent dual chamber pacemaker was significantly higher than that of patients with a permanent ventricular pacemaker (p < 0.02).

~ 100 Q) .... as a:: iii 80 > 'e :::I 60 en Q) .! .. as 40 :i E :::I 0 20 ..... Ventricular-paced; n = 53 1) 0--0 Dual chamber-paced; n = 20 Q) ..

p< 0.02 u :g ! 0 Q. 1 2 3 4 5

Years Following Pacemaker Implantation

lACC Vol 7. No 4 Apnl 1986925-32

Table 5. Causes of Death

Ventricular Paced Cause of Death (Group I)

Total 132 Total dead 45 Cause of death known 40

Acute myocardial IS infarction

Congestive heart 13 failure

Cancer 4 Pneumonia 2 Stroke 2 Pulmonary emboli 2 Emphysema/chronic

bronchitis Renal failure I

Cause of death unknown 5 Sudden; pacemaker

function normal Sudden; pacemaker 2

function unknown Nonsudden 2

Dual Chamber Paced (Group 2)

48 7 7 3

o o o

I o o

o

o

likely to impair cardiac performance in patients with preex•istent left ventricular dysfunction. Properly timed atrial or dual chamber pacing restores atrial transport function, re•sulting in improved cardiac performance (23-30). Using data obtained during ventricular pacing as a baseline, Stew•art et al. (31) demonstrated that dual chamber pacing produced an average overall increase of 19% in cardiac output (mea•sured using continuous wave Doppler recording) and a 30% increase in patients with 1: 1 ventriculoatrial VVI pacing. Ditsch et al. (32) observed a 53.5% rise in cardiac index and a 7.5% decrease in left ventricular end-diastolic volume in 16 patients with normal systolic left ventricular function at rest during the transition from VVI to DDD pacing. Reiter and Hindman (33) studied seven patients with depressed left ventricular ejection fraction (mean 29%) and New York Heart Association functional class III or IV congestive heart failure. Cardiac index was higher in all patients at heart rates of 75 to 100 beats/min with A V sequential pacing compared with VVI pacing. The mean improvement in car•diac index was 17% at a heart rate of 75 beats/min, 23% at a rate of 85 beats/min and 29% at a rate of 100 beats/min. A recent study by DiCola et al. (34) demonstrated greater exercise tolerance with dual chamber pacing than with ven•tricular pacing, suggesting that the comparative hemody•namic benefits of dual chamber pacing might result in clin•ical benefits.

Comparative survival after permanent ventricular and dual chamber pacing in patients with and without preex•istent congestive heart failure. Our study is the first to compare survival characteristics of patients with chronic

ALPERT ET AL SURVIVAL AFTER PERMANENT PACEMAKER IMPLANTATION

931

high degree A V block receiving a permanent ventricular pacemaker with those of patients receiving a dual chamber pacemaker for chronic high degree A V block. It is also the first study to explore the comparative impact of congestive heart failure on survival in these two groups. Our results show that permanent dual chamber pacing (Group 2) en•hances survival (relative to permanent ventricular pacing, Group I) in patients with chronic high degree A V block and preexistent congestive heart failure, but not in patients without preexistent failure. There were no significant dif•ferences in predicted cumulative survival rates between Groups 1 and 2 for patients with potential underlying causes of congestive heart failure (hypertension, coronary artery dis•ease, valvular heart disease). This suggests that enhanced survival in Group 2 may be the result of the comparative hemodynamic advantages of dual chamber pacing.

Limitations. Our conclusions should be interpreted in light of certain limitations imposed by the study design. Because this study was retrospective, we were forced in many cases to rely on clinical information obtained by phy•sicians other than those directly involved in the study. In particular, left ventricular function was not routinely mea•sured. This limitation may have influenced our ability to accurately diagnose left ventricular dysfunction and cer•tainly impaired our ability to gauge its severity. To counter this, we required patients to demonstrate criteria for definite heart failure as delineated in the Framingham study for inclusion in the congestive heart failure subgroups. Com•parison of subgroups in a retrospective study is hazardous because of the potential dissimilarity of patient character•istics and treatment modalities between groups. As shown in Tables 1 to 3, the patient characteristics and incidence of coexistent illnesses in Groups 1 and 2 were similar (the incidence of coronary artery disease, however, was actually higher in the dual chamber paced group than in the ven•tricular paced group). Moreover, there was no significant difference between the two groups in the frequency of use of medical or surgical treatment modalities for the most common coexistent illnesses (Table 4). This suggests that Groups 1 and 2 were comparable (although not matched) despite the long interval of time permitted for entry of patients.

Conclusion. Despite the potential limitations of this study, we believe the results suggest that permanent dual chamber pacing enhances survival to a greater extent than does per•manent ventricular pacing in patients with chronic high de•gree A V block and preexistent congestive heart failure.

We thank Cindy Branson for secretarial assistance and Rebecca J. Morgan, RN for technical assistance.

References I. Johansson BW. Longevity in complete heart block. Ann NY Acad

Sci 1969;167:1031-7.

932 ALPERT ET AL SURVIVAL AFfER PERMANENT PACEMAKER IMPLANTATION

2. Rowe JC, White PD. Complete heart block: a follow-up study. Ann Intern Med 1957;49:260-75.

3. Pader E, Levy H. Clinical and electrocardiographic studies in complete heart block. J Chronic Dis 1966;19:1101-22.

4. Bernstein V, Rostem CE, Peretez DJ. Permanent pacemakers: eight year follow-up study. Ann Intern Med 1971 ;74:361-9.

5. Amikam S, Lerner J, Roglun V, et a!. Long-term survival of elderly patients after pacemaker implantation. Am Heart J 1976;9:445-9.

6. Seremetis MG, de Guzman VC, Lyons S, et a!. Cardiac pacemakers: clinical experience with 289 patients. Am Heart J 1973;85:739-48.

7. Davidson DM, Braall CA, Preston TA, et a!. Permanent ventricular pacing: effect on long term survival, congestive heart failure and subsequent myocardial infarction and stroke. Ann Intern Med 1972;77:345-51.

8. Zion MM, Marchand PE, Obel IWP. Long-term prognosis after car•diac pacing in atrioventricular block. Br Heart J 1973;35:359-64.

9. Lagergren H, Johansson L, LandergrenJ, et a!. 305 cases of permanent pacemaker treatment for Adams-Stokes syndrome. Surgery 1966;59:494-7.

10. Sowton E, Flores J. Natural history of pacemaker patients. Bull NY Acad Med 1971;47:999-1010.

II. Hansen J, Meibom J. The prognosis for patients with complete heart block treated with permanent pacemaker. Acta Med Scand 1974;195: 385-9.

12. Torresam J, Bernard Y, Monties JR, et a!. Clinical experience in transvenous and myocardial pacing. Ann NY Acad Sci 1969;167:996-1007.

13. Alpert MA, Katti SK. Natural history of high-degree atrioventricular block following permanent pacemaker implantation. J Chronic Dis 1982;35:341-4.

14. Hauser RG, Jones 11, Moss K, et a!. SurvIval after pacemaker im•plantation. In: Proceedings of the 7th World Symposium on Cardiac Pacing. Vienna: Durmstadt, Stein Kapff Verlag, 1983:483-90.

15. Simon AB, Janz N. Symptomatic bradyarrhythmias in the adult: nat•ural history following ventricular pacemaker implantation. PACE 1982;5:372-83.

16. McKee PA, Castelli WP, McNamara R. The natural history ofconges•tive heart failure. The Framingham study. N Engl J Med 1971;285:1441-6.

17. Gehan EA. Estimating survival functions from the life table. J Chronic Dis 1969;21 :629-44.

18. Breslow N. A generalized Kruskal-Wallis test for comparing K sam•ples subject to unequal patterns of censorship. Biometrika 1970;57:579-86.

JACC Vol. 7, No 4 Apnl 1986.925-32

19. Laczkovics A, Mohl W, Stembach K. Life expectancy of patients in relation to age and indications for pacemaker treatment. In: Meere C, ed. Proceedings of the 6th World Symposium on Cardiac Pacmg. Montreal. 1979; 13-4.

20. Castberg T. PatIent survival during pacemaker treatment. Scand J Thorac Cardiovasc Surg (Suppl) 1978;22:57-62.

21. Chamberland DA, Leinbach RC, Vassauz CEo Sequential atrioven•tricular pacing in heart block complicating acute myocardial infarction. N Engl J Med 1970;282:577-82.

22. Hartzler GO, Maloney JD, Curtis 11, et a!. Hemodynamic benefits of A V sequential pacing after cardiac surgery. Am J Cardiol 1977;40:232-6.

23. Hayes DL, Holmes DR Jr, Vlietstra RE, Osborn MJ. Changing ex•perience with dual chamber (DDD) pacemakers. J Am Coli Cardiol 1984;4:556-9.

24. Parsonnet V, Bernstein AD. CardIac pacing in the 1980s. Treatment and techniques in tranSItion. J Am Coli Cardiol 1983;1 :339-54.

25. Sutton R, Perrins J, Atron P. Physiologic cardiac pacing. PACE 1980;3:207-19.

26. Hawthorne JW. The physiology of cardiac pacing. Intelligence Rep Cardiac Pacing..-Electrophysiol 1984;3: 1-7.

27. Judge RD, Wilson WS, Seigel JH. Hemodynamic studies in patients with cardiac pacemakers. N Engl J Med 1964;21:1391-5.

28. Samet PS, Castillo C, Bernstein WHo Hemodynamic consequences of sequential atrioventricular pacing. Am J Cardiol 1968;21:7-12.

29. Benchimol A, Ellis 10, Dimond EG. Hemodynamic consequences of atrial and ventricular pacing in patients with normal and abnormal hearts. Am J Med 1968;39:911-22.

30. Kruse I, Amman K, Conradson TB, Ryden L. A comparison of the acute and long-term hemodynamics of ventricular inhibited and atrial synchronous ventricular inhibited pacing. Circulation 1982;65:846-55.

31. Stewart WJ, DiCola VL, Hawthorne JW. Doppler ultrasound mea•surement of cardiac output in patients with physiologic pacemakers: effects of left ventricular function and retrograde ventriculoatrial con•duction. Am J Cardiol 1984;54:308-12.

32. Ditsch J, Seiderer M, Bull U, Luderitz W. Evaluation of left ven•tricular performance in patients with atrioventricular versus ventricular demand pacemakers. Am Heart J 1984;107:906-11.

33. Reiter MJ, Hindman MC. Hemodynamic effects of acute AV se•quential pacing in patients with left ventricular dysfunction. Am J CardlOl 1982;49:687-92.

34. DiCola VC, Hand R, Boucher CA. Exercise cardiopulmonary as•sessment with dual chamber versus ventricular pacing. PACE 1983;6:A-81.