Incidence and Predictors of Loss of Pacing in the Atrium in Patients with Sick Sinus Syndrome ELENA B. SGARBOSSA, SERGIO L. PINSKI, LON W. GASTLE, RIGHARD G. TROHMAN, and JAMES D. MALONEY From the Department of Cardiology, The Cleveland Clinic Foundation, Cleveland, Ohio
SGARBOSSA, E.B., ET AL.: Incidence and Predictors of Loss of Pacing in the Atrium in Patients with Sick
Sinus Syndrome. AtriaJ and dual-chamber pacemakers may be associated with reduced morbidity in patients ivith the sick sinus syndrome (SSS). In some patients, however, subsequent development of chronic atrial fibrillation or atrial lead failure make long-term pacing in the atrium not feasible. We analyzed the incidence and predictors of loss of atrial pacing in 395 consecutive patients with SSS (376 with dual-chamber pacemakers and 19 with single-chamber atriaJ pacemakers). None of them was in estabJished atriaJ fibrillation at time of implant. Patients were foJlowed-up for 55 ± 35 months. Actuarial survival of effective atrial pacing was 92.5% at 1 year, 85% at 5 years, and 76.5% at 10 years. Overall, 60 patients lost atrial pacing. The most frequent cause was the development of chronic atrial fibrillation (53 patients]. By muJtivariate analysis (Cox proportional-hazards model), independent predictors of Joss of pacing in the atrium were preimplant episodes of paroxysmal atrial fibrillation (PAF) lasting more than 1 hour fP < 0.001; hazard ratio fHR) = 4.3); prior history of PAF for more than 5 years (P < 0.001 ; Hfl = 2.67; and endocardial P wave < 2 mV (P = 0.014; HR = 1.96). In a subgroup of patients fn = 187J who had echocardiograms, a left atrium > 50 mm was also an independent predictor of loss of atriai pacing (P = 0.028; HR = 2.28). Conclusions: 1) most patients with SSS can maintain Jong-term atcial pacing; 2) loss of pacing in the atrium is related to the previous history of PAF, left atrial enlargement, and low amplitude of the endocardial P wave at implant; and 3) patients with these risks variables are Jess than ideal candidates for atrial pacing modes. In them, the implant of DDDR units might be indicated, to provide wide fJexibiJity in case reprogramming to a ventricuJar pacing mode is required. (PACE, VoJ. 15, November, Part II 1992) sick sinus syndrome, dual-chamber pacing, complications
Introduction It has been reported that atrial and dual-chamber pacing in sick sinus syndrome (SSS) are associated with decreased morbidity.^ In some patients, however, development of chronic atrial fibrillation or atrial lead failure make long-term pacing in the atrium not feasible. The higher costs of dualchamber devices,^ or the costs of a reintervention in case of failure of a single-chamber atrial pace-
Address for reprints: Elena B. Sgarbossa, M.D., Dept. of Cardiology, Desk F 15, The Cleveland Clinic Foundation, 9500 Euclid Ave., Cleveland, OH 44195. Fax: (216) 444-0456.
maker might be prevented if identification a priori of these patients were possible. To analyze the incidence and predictors of loss of pacing in the atrium we performed a long-term follow-up on 395 patients with SSS. Methods Study Patients
Between January 1980 and December 1989, we implanted 19 atrial and 376 dual-chamber pacemakers in 395 adult patients (age more than 18 years) with SSS. SSS was defined by the presence of inappropriate, persistent sinus bradycar-
Novemher, Part II 1992
PACE, Vol. 15
SURVIVAL OF ATRIAL PACING MODES
dia (rate < 50 beats/min), sinus pauses longer than 3 seconds, or sinoatrial block. No patient was in established atrial fibrillation at time of pacemaker implant. Patients with concomitant complete AV block or type II second degree AV block were not included in the study. All patients were symptomatic, required bradycardia-producing drugs for treatment of tachyarrhythmias, or both. Seventeen baseline variables describing cardiac and concomitant diseases, electrocardiographic and echocardiographic findings were analyzed (Table I). The clinical characteristics of preimplant paroxysmal atrial fibrillation (PAF) were the object of particular study. Time since first documented episode of PAF varied from 1 to 485 months. For the purposes of this analysis the population was dichotomized upon a history of PAF shorter or longer than 5 years. From results of ambulatory Holter recordings, in-hospital telemetry, and self-assessment by symptomatic patients, the duration of the longest episode of PAF in each patient was classified as lasting less or more than 1 hour. Need for electrical cardioversion and/or chronic treatment with antiarrhythmic drugs prior to pacemaker implant were also recorded (Table I). The voltage of the endocardial P wave at time of implant was dichotomized as being equal or greater versus < 2.0 mV. Follow-Up Loss of pacing in the atrium for at least 3 consecutive months was the endpoint in this study. Patients were foUowed-up for a mean of 55 ± 35 months (range: 1-97 months). Follow-up began on the date of pacemaker implant, and ended on the date of the atrial lead failure or abandonment, patient's death, or the end of the study (December 31, 1990). Survival status, underlying cardiac rhythm, generator or lead replacement, and eventual reprogramming of the initial pacing mode were ascertained through review of medical records, questionnaires completed by the patient's private physician, and telephonic interviews with the patients or their families. Follow-up was complete in 98.5% of the patients. Eight patients were lost to follow-up after a mean of 56 months and had their follow-up censored at time of last contact. Statistical Analysis Continuous variables are presented as mean ± 1 SD. Actuarial curves for maintenance of atrial
PACE, Vol. 15
Table 1. Clinical Characteristics Clinical variables Age, years Male/Female
66 ± 12 248/147
Structural heart disease* Coronary artery disease Valvular heart disease Cardiomyopathy Other
234 (59%) 173 (44%) 45 (12%) 29 (7%) 40 (10%)
Concomitant diseases Hypertension Diabetes Cerebrovascular disease Peripheral vascular disease
161 (14%) 54 (14%) 11 (3%) 45 (7%)
Electrocardiographic findings Bundle branch block Paroxysmal atrial tachyarrhythmia Long history of PAF (> 5 years) Prolonged episodes of PAF (> 1 hour) Complex ventricular arrhythmia
59 (15%) 298 (74%) 46 (12%) 92 (24%) 136 (34%)
Medication Preimplant antiarrhythmic drugs Preimplant digoxin Discharge antiarrhythmic drugs Discharge digoxin
69 (17%) 187 (47%) 125 (32%) 119 (30%)
Left atrial enlargementt Preimplant electrical cardioversion Endocardial P voltage a 2.0 mV Endocardial P voltage < 2.0 mV P voltage not recorded
92 (50%) 18 (5%) 298 (75%) 83 (21%) 14
* Some patients had more than one cardiac disease; 1 1 187.
pacing and for each variable were calculated with the method of Kaplan and Meier. Analyzed variables were screened by univariate statistical methods to identify those associated with loss of atrial pacing. Multivariate regression analysis (Cox proportional-hazard model), was applied to all variables that had at least marginal univariate predictive value (P < 0.10). This analysis was used to identify variables with significant independent predictive value (P < 0.05).
November, Part II 1992
SCARBOSSA, ET AL.
Table III. Univariate Predictors of Loss of Atrial Pacing Variables
Figure 1. Actuarial survival of pacing in the atrium.
Results Sixty patients (15%) had loss of atrial pacing during the follow-up period. Actuarial survival of effective pacing in the atrium was 92.5% at 1 year, 85% at 5 years, and 76.5% at 10 years (Fig. 1). Among the 60 patients who lost atrial pacing, 4 had AAI pacemakers and 56 had a dual-chamber system (DDD, DVI, or DDDR). The most frequent cause of loss of pacing in the atrium was the development of chronic atrial fibrillation, present in 53 patients (two with AAI and 51 with dual-chamber pacemakers). Causes other than chronic atrial fibrillation are listed in Table II. The clinical predictors of loss of pacing in the
Table II. Causes of Loss of Atrial Pacing Other Than Chronic Atrial Fibrillation Age
Competitive pacing leading to PAF Exit block Frequent episodes of
Frequent episodes of
History of PAF for more than 5 yrs Episodes of PAF lasting more than 1 hour Valvular heart disease Preimplant digoxin Age Endocardial P wave < 2.0 mV Preimplant antiarrhythmic drugs
4.3 4.1 NA
Atrial undersensing Atrial undersensing Exit block
PAF = paroxysmal atrial fibrillation.
0.001 0.001 0.001 0.001 0.008 0.02 0.03
atrium by univariate analysis are summarized in Table III. Survival curves for effective atrial pacing according to duration of the longest preimplant episode of atrial fibrillation, the prior history of PAF, and the voltage of the endocardial P wave are depicted in Figures 2, 3, and 4, respectively. The independent predictors of loss of pacing in the atrium were episodes of PAF lasting more than 1 hour (P < 0.001; hazard ratio (HR) = 4.3; confidence interval (Cl) = 2.79 to 6.91); history of PAF longer than 5 years (P < 0.001; HR = 2.67; Cl = 1.51 to 4.71), and endocardial P wave < 2 mV (P = 0.014; HR = 1.96; Cl = 1.14 to 3.36). Among the subgroup of 187 patients who had echocardiographic assessment prior to pacemaker implant, independent predictors were episodes of PAF lasting more than 1 hour (P < 0.001; HR = 4.2; Cl = 1.00
PAF 72 73 74
< < <