Pacemaker Dependency After Coronary Artery Bypass SHLOMO FELDMAN, MICHAEL GLIKSON, and ELIESER KAPLINSKY From the Heart Institute, Ghaim Sheba Medical Genter and the Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
FELDMAN, S., ET AL.: Paceinaker Dependency After Coronary Artery Bypass. A retrospective study was
carried out on 36 patients (33 males and 3 females) to determine the incidence of WI pacemaker dependency following coronary artery bypass surgery. Pacemaker dependency was defined as the presence of pacemaker activity when pacing rate was programmed at 50 beats/min and/or when no hemodynamic adequate rhythm was present during puJse generator replacement. The patients were divided into two groups: (group Ij 26 patients with complete atrioventricular (AV) block developing in the early postoperative period. In most of them a pacemaker was implanted up to 3 weeks following surgery (range 2 days to 1 year); (group II) ten patients in whom the indication for pacemaker implantation was sick sinus syndrome with sinus arrest and/or tachy-bradycardia. These patients underwent pacemaker implantation at varying periods of time following surgery (range 12 days to 4 years'). Unipolar endocardial leads with W I programmable pacemakers were implanted in aJJ patients included in this study. Mean foJJow-up time was 3 years. In group I the pacemaker dependency rate was 65%, whereas in group II it was 30% throughout the foJlow-up period. It is concluded that the Jow incidence of pacemaker dependency in patients who undergo pacemaker implantation after coronary bypass surgery necessitates frequent evaluation in the nondependent patient, in order to reassess the need for the pacemaker before puJse generator replacement. Such reassessment should probably include prolonged ambulatory monitoring as well as invasive evaluation of the conduction system, if avoidance of puJse generator replacement is considered. (PACE, Vol. 15, November, Part II 1992J pacemaker dependency, coronary artery bypass
Introduction A selective subset of patients after coronary artery bypass surgery underwent permanent pacemaker implantation because of postoperative complete AV block, sinus arrest, or tacby-bradycardia syndrome. A retrospective study was carried out to determine the incidence of pacemaker dependency during long-term follow-up. Materials and Methods Thirty-six patients (33 males and 3 females) were included in this retrospective study. Two
Address for reprints: Shlomo Feldman, M.D., Tel Aviv University, Tel Hashomer, 52621 Tel Aviv, Israel.
PAGE, Vol. 15
groups of patients were selected according to the indication for pacemaker implantation. Group I: 26 patients with postoperative complete AV block; and group II: 10 patients with sick sinus syndrome, presenting as sinus arrest and/or tachybradycardia. Patient data are presented in Table I. The only significant differences between the two groups were the incidence of diabetes mellitus, and the incidence of preoperative conduction disturbances. Time of implantation was 3 weeks or less following surgery in most of the patients in group I (only four underwent implantation between 3 weeks and 3 months, and only one later than this). Time of implantation varied considerably in the second group, and was generally longer than in the first group (range 12 days to 4 years). Implantation in five patients in this group was performed
November, Part II 1992
2037
FELDMAN, ET AL.
Table 1. Patient Data Group 1 (26 Patients)
Group II (10 Patients)
P*
Indications tor impiantation
Complete AV block
Time of implantation Males/Females Old myocardial infarction Diabetes mellitus Mitral valve replacement** Aortic valve replacement** Average number of grafts Conduction disturbances prior to surgery^
2 days to 1 year 25/1 14 (54%) 23 (88%) 3 (12%) 2 (8%) 3.3 ± 1.3 13 (50%)
Sinus arrest (7) Tachy-bradycardia (3) 12 days to 5 years 8/2 5 (50%) 0 0 0 3.9 ± 0.8 0
NS/NS NS < 0.0001 NS NS NS < 0.005
* P = difference between two groups (Fisher's exact test and f-test); ** during bypass surgery; f conduction disturbances including right or left bundle branch block, first-degree AV block, and bifascicular and trifascicular block. NS = nonsignificant difference (P > 0.05).
more than 1 year following surgery. Patients with DDD pulse generators, recent implantation (where follow-up was too short), and patients on medication with negative chronotropic and/or dromotropic effects were excluded. Mean follow-up time was 3 years. Pacemaker dependency was defined as the presence of pacemaker activity when the pacing rate was programmed at 50 beats/min and/or when no hemodynamic adequate rhythm was present during pulse generator replacement. Distribution of variables in the two groups was compared using Fisher's exact test, and averages were compared by t-test. Results The pacemaker dependency rate in the two groups is presented in Table II. Whereas 65% of
Table II.
the patients with postoperative complete AV block were found dependent, only 30% of the patients in the second group, with other bradyarrhythmias, were dependent during their follow-up period. The difference between the groups, however, did not reach statistical significance. Table III compares the two subgroups of dependent versus nondependent patients with complete AV block all belonging to group I. Group II was too small for any statistical analysis. No difference could be found between the two subgroups (dependent vs nondependent) regarding age, sex, number of grafts, preoperative ejection fraction, presence of preoperative conduction disturbances, old myocardial infarction, and the average time of implantation. Nondependent patients tended to undergo implantation somewhat later; most of them during the third postoperative week, while most of the dependent patients underwent implantation during the first 2 weeks following surgery (Table IV).
Pacemaker Dependency
Group 1 (26 patients) Group II (10 patients)
Discussion
No. of Patients
%
P*
17 3
65 30
NS NS
* P is difference between groups (Fisher's exact test).
2038
Postoperative complete AV block or other severe bradyarrhj^hmias occur in about 4% of patients undergoing coronary bypass surgery. ^-^ Risk factors for this complication include multiple vessel disease,^ aortic cross-clamping and pump
November, Part II 1992
PACE, Vol. 15
PACEMAKER DEPENDENCY AFTER CORONARY ARTERY BYPASS
Table III. Pacemaker Dependency Among Group I Patients
Age (years) Males/Females Number of grafts Implantation time (days) Ejection fraction prior to surgery Conduction disturbances prior to surgery Old myocardial infarction
Dependent (17 Patients)
Nondependent (9 Patients)
66.4 ± 5.9 16/1 3.5 ± 1.1 20.9 ± 25.5** (2 days-3 months) 45.3 ± 14.3 7 9
67.1 ± 4.8 9/0 2.9 ± 1.4 24.8 ± 23.5 (5 days-3 months) 41.2 ± 12.0 6 5
NS NS/NS NS NS NS NS NS
* P is difference between groups (/-test and Fisher's exact test); ** in calculating this average we ignored a single case of implantation at 1 year. NS = nonsignificant.
^ old age,^ and preexisting bundle branch block. ^ In addition, certain features of the preoperative coronary anatomy have been correlated to severe postoperative conduction disturbances.^'^ Factors associated with the constituents and temperature of the cardioplegic solution may also affect conduction during the early postoperative period.^'^ It has been clearly established in the past^ that postoperative AV block may be reversible either during the first few days after the operation or during the first few weeks following hospital discharge. This is probably due to the role of transient factors such as ischemia without necrosis and local edema, which create these postoperative conduction system disturbances. This is supported by our findings in group I, in which 35% of the patients were nondependent 3 years after
Tabie iV. Time of Implantation (Group I) Dependent (17 Patients) < 1 week
1-2 weeks 2-3 weeks 3 weeks-3 months > 3 months
PACE, Vol. 15
4 7 2 3 1 (1 year)
Nondependent (9 Patients) 1 1 6 1
surgery. We could not find any correlation between nondependency and other pre- or postoperative factors, thereby limiting the ability to identify prospectively non-dependent patients. Nevertheless, we believe that reassessment of the need for a pacemaker should be carried out in all postoperative patients prior to pulse generator replacement, since up to 35% of them may no longer require a pacemaker. The low rate of dependency in group II may be an expression of the more liberal indications for implantation in this group, as well as the possibly more impredictable course of the bradyarrhythmias, with intermittent pacemaker dependency, which may not have been readily detected by our criteria. Nevertheless, reassessment of the need should be considered before pulse generator replacement in this group also. This reassessment should at least include prolonged ambulatory monitoring while the pacemaker is programmed at < 50 beats/min, and invasive electrophysiological assessment of sinus node, as well as AV nodal function. Unfortunately, this group was too small for us to reach conclusions regarding correlations between risk factors and long-term dependency. Therefore, we could not identify prospectively patients destined to become pacemaker nondependent. In conclusion, many patients with postoperative conduction disturbances, especially of the sick sinus syndrome type, may not be pacemaker
November, Part II 1992
2039
FELDMAN, ET AL.
dependent on long-term follow-up, as may some of the patients with complete AV block. We, therefore, believe that all these patients should be monitored frequently for pacemaker dependency, with comprehensive reassessment of their status by
prolonged ambulatory and invasive electrophysiological evaluation prior to pulse generator replacement, with the aim of avoiding unnecessary replacement.
References 1. Baerman JM, Kirsh MM, De Buitleir M, et al. Natural history and determinants of conduction defects following coronary artery bypass surgery. Ann Thorac Surg 1987; 44:150-153. 2. Moore SL, Wilkoff BL. Rhythm disturbances after cardiac surgery. Semin Thorac Cardiovasc Surgery 1991; 3:24-28. 3. Emlein G, Rofino K, Mittleman RS, et al. Severe bradyarrh5^hmias requiring permanent pacemaker insertion after coronary artery bypass surgery: Incidence, clinical and electrocardiographic characteristics, (abstract) PACE 1992; 15:508. 4. Caspi J, Amar R, Elami A, et al. Frequency and significance of complete atrioventricular block after
2040
coronary artery bypass grafting. Am J Ccirdiol 1989; 63:526-529. 5. Mosseri M, Meir C, Lotan C, et al. Coronary pathology predicts conduction disturbances after coronary artery bypass grafting. Am Thorac Surg 1991; 51: 248-252. 6. Rippe JM, Browning C, Vander Salm T, et al. Fascicular conduction disturbances following aorto-coronary bypass surgery: The role of hypothermia versus potassium arrest cardioplegia. J Cardiovasc Surg 1984; 25:456-461. 7. Gundry SR, Sequeira A, Coughlin TR, et al. Postoperative conduction disturbances: A comparison of blood and crystalloid cardioplegia. Am Thorac Surg 1989; 47:384-390.
November, Part n 1992
PACE, Vol. 15
FELDMAN, S., ET AL.: Paceinaker Dependency After Coronary Artery Bypass. A retrospective study was
carried out on 36 patients (33 males and 3 females) to determine the incidence of WI pacemaker dependency following coronary artery bypass surgery. Pacemaker dependency was defined as the presence of pacemaker activity when pacing rate was programmed at 50 beats/min and/or when no hemodynamic adequate rhythm was present during puJse generator replacement. The patients were divided into two groups: (group Ij 26 patients with complete atrioventricular (AV) block developing in the early postoperative period. In most of them a pacemaker was implanted up to 3 weeks following surgery (range 2 days to 1 year); (group II) ten patients in whom the indication for pacemaker implantation was sick sinus syndrome with sinus arrest and/or tachy-bradycardia. These patients underwent pacemaker implantation at varying periods of time following surgery (range 12 days to 4 years'). Unipolar endocardial leads with W I programmable pacemakers were implanted in aJJ patients included in this study. Mean foJJow-up time was 3 years. In group I the pacemaker dependency rate was 65%, whereas in group II it was 30% throughout the foJlow-up period. It is concluded that the Jow incidence of pacemaker dependency in patients who undergo pacemaker implantation after coronary bypass surgery necessitates frequent evaluation in the nondependent patient, in order to reassess the need for the pacemaker before puJse generator replacement. Such reassessment should probably include prolonged ambulatory monitoring as well as invasive evaluation of the conduction system, if avoidance of puJse generator replacement is considered. (PACE, Vol. 15, November, Part II 1992J pacemaker dependency, coronary artery bypass
Introduction A selective subset of patients after coronary artery bypass surgery underwent permanent pacemaker implantation because of postoperative complete AV block, sinus arrest, or tacby-bradycardia syndrome. A retrospective study was carried out to determine the incidence of pacemaker dependency during long-term follow-up. Materials and Methods Thirty-six patients (33 males and 3 females) were included in this retrospective study. Two
Address for reprints: Shlomo Feldman, M.D., Tel Aviv University, Tel Hashomer, 52621 Tel Aviv, Israel.
PAGE, Vol. 15
groups of patients were selected according to the indication for pacemaker implantation. Group I: 26 patients with postoperative complete AV block; and group II: 10 patients with sick sinus syndrome, presenting as sinus arrest and/or tachybradycardia. Patient data are presented in Table I. The only significant differences between the two groups were the incidence of diabetes mellitus, and the incidence of preoperative conduction disturbances. Time of implantation was 3 weeks or less following surgery in most of the patients in group I (only four underwent implantation between 3 weeks and 3 months, and only one later than this). Time of implantation varied considerably in the second group, and was generally longer than in the first group (range 12 days to 4 years). Implantation in five patients in this group was performed
November, Part II 1992
2037
FELDMAN, ET AL.
Table 1. Patient Data Group 1 (26 Patients)
Group II (10 Patients)
P*
Indications tor impiantation
Complete AV block
Time of implantation Males/Females Old myocardial infarction Diabetes mellitus Mitral valve replacement** Aortic valve replacement** Average number of grafts Conduction disturbances prior to surgery^
2 days to 1 year 25/1 14 (54%) 23 (88%) 3 (12%) 2 (8%) 3.3 ± 1.3 13 (50%)
Sinus arrest (7) Tachy-bradycardia (3) 12 days to 5 years 8/2 5 (50%) 0 0 0 3.9 ± 0.8 0
NS/NS NS < 0.0001 NS NS NS < 0.005
* P = difference between two groups (Fisher's exact test and f-test); ** during bypass surgery; f conduction disturbances including right or left bundle branch block, first-degree AV block, and bifascicular and trifascicular block. NS = nonsignificant difference (P > 0.05).
more than 1 year following surgery. Patients with DDD pulse generators, recent implantation (where follow-up was too short), and patients on medication with negative chronotropic and/or dromotropic effects were excluded. Mean follow-up time was 3 years. Pacemaker dependency was defined as the presence of pacemaker activity when the pacing rate was programmed at 50 beats/min and/or when no hemodynamic adequate rhythm was present during pulse generator replacement. Distribution of variables in the two groups was compared using Fisher's exact test, and averages were compared by t-test. Results The pacemaker dependency rate in the two groups is presented in Table II. Whereas 65% of
Table II.
the patients with postoperative complete AV block were found dependent, only 30% of the patients in the second group, with other bradyarrhythmias, were dependent during their follow-up period. The difference between the groups, however, did not reach statistical significance. Table III compares the two subgroups of dependent versus nondependent patients with complete AV block all belonging to group I. Group II was too small for any statistical analysis. No difference could be found between the two subgroups (dependent vs nondependent) regarding age, sex, number of grafts, preoperative ejection fraction, presence of preoperative conduction disturbances, old myocardial infarction, and the average time of implantation. Nondependent patients tended to undergo implantation somewhat later; most of them during the third postoperative week, while most of the dependent patients underwent implantation during the first 2 weeks following surgery (Table IV).
Pacemaker Dependency
Group 1 (26 patients) Group II (10 patients)
Discussion
No. of Patients
%
P*
17 3
65 30
NS NS
* P is difference between groups (Fisher's exact test).
2038
Postoperative complete AV block or other severe bradyarrhj^hmias occur in about 4% of patients undergoing coronary bypass surgery. ^-^ Risk factors for this complication include multiple vessel disease,^ aortic cross-clamping and pump
November, Part II 1992
PACE, Vol. 15
PACEMAKER DEPENDENCY AFTER CORONARY ARTERY BYPASS
Table III. Pacemaker Dependency Among Group I Patients
Age (years) Males/Females Number of grafts Implantation time (days) Ejection fraction prior to surgery Conduction disturbances prior to surgery Old myocardial infarction
Dependent (17 Patients)
Nondependent (9 Patients)
66.4 ± 5.9 16/1 3.5 ± 1.1 20.9 ± 25.5** (2 days-3 months) 45.3 ± 14.3 7 9
67.1 ± 4.8 9/0 2.9 ± 1.4 24.8 ± 23.5 (5 days-3 months) 41.2 ± 12.0 6 5
NS NS/NS NS NS NS NS NS
* P is difference between groups (/-test and Fisher's exact test); ** in calculating this average we ignored a single case of implantation at 1 year. NS = nonsignificant.
^ old age,^ and preexisting bundle branch block. ^ In addition, certain features of the preoperative coronary anatomy have been correlated to severe postoperative conduction disturbances.^'^ Factors associated with the constituents and temperature of the cardioplegic solution may also affect conduction during the early postoperative period.^'^ It has been clearly established in the past^ that postoperative AV block may be reversible either during the first few days after the operation or during the first few weeks following hospital discharge. This is probably due to the role of transient factors such as ischemia without necrosis and local edema, which create these postoperative conduction system disturbances. This is supported by our findings in group I, in which 35% of the patients were nondependent 3 years after
Tabie iV. Time of Implantation (Group I) Dependent (17 Patients) < 1 week
1-2 weeks 2-3 weeks 3 weeks-3 months > 3 months
PACE, Vol. 15
4 7 2 3 1 (1 year)
Nondependent (9 Patients) 1 1 6 1
surgery. We could not find any correlation between nondependency and other pre- or postoperative factors, thereby limiting the ability to identify prospectively non-dependent patients. Nevertheless, we believe that reassessment of the need for a pacemaker should be carried out in all postoperative patients prior to pulse generator replacement, since up to 35% of them may no longer require a pacemaker. The low rate of dependency in group II may be an expression of the more liberal indications for implantation in this group, as well as the possibly more impredictable course of the bradyarrhythmias, with intermittent pacemaker dependency, which may not have been readily detected by our criteria. Nevertheless, reassessment of the need should be considered before pulse generator replacement in this group also. This reassessment should at least include prolonged ambulatory monitoring while the pacemaker is programmed at < 50 beats/min, and invasive electrophysiological assessment of sinus node, as well as AV nodal function. Unfortunately, this group was too small for us to reach conclusions regarding correlations between risk factors and long-term dependency. Therefore, we could not identify prospectively patients destined to become pacemaker nondependent. In conclusion, many patients with postoperative conduction disturbances, especially of the sick sinus syndrome type, may not be pacemaker
November, Part II 1992
2039
FELDMAN, ET AL.
dependent on long-term follow-up, as may some of the patients with complete AV block. We, therefore, believe that all these patients should be monitored frequently for pacemaker dependency, with comprehensive reassessment of their status by
prolonged ambulatory and invasive electrophysiological evaluation prior to pulse generator replacement, with the aim of avoiding unnecessary replacement.
References 1. Baerman JM, Kirsh MM, De Buitleir M, et al. Natural history and determinants of conduction defects following coronary artery bypass surgery. Ann Thorac Surg 1987; 44:150-153. 2. Moore SL, Wilkoff BL. Rhythm disturbances after cardiac surgery. Semin Thorac Cardiovasc Surgery 1991; 3:24-28. 3. Emlein G, Rofino K, Mittleman RS, et al. Severe bradyarrh5^hmias requiring permanent pacemaker insertion after coronary artery bypass surgery: Incidence, clinical and electrocardiographic characteristics, (abstract) PACE 1992; 15:508. 4. Caspi J, Amar R, Elami A, et al. Frequency and significance of complete atrioventricular block after
2040
coronary artery bypass grafting. Am J Ccirdiol 1989; 63:526-529. 5. Mosseri M, Meir C, Lotan C, et al. Coronary pathology predicts conduction disturbances after coronary artery bypass grafting. Am Thorac Surg 1991; 51: 248-252. 6. Rippe JM, Browning C, Vander Salm T, et al. Fascicular conduction disturbances following aorto-coronary bypass surgery: The role of hypothermia versus potassium arrest cardioplegia. J Cardiovasc Surg 1984; 25:456-461. 7. Gundry SR, Sequeira A, Coughlin TR, et al. Postoperative conduction disturbances: A comparison of blood and crystalloid cardioplegia. Am Thorac Surg 1989; 47:384-390.
November, Part n 1992
PACE, Vol. 15
