Summary
Background
The prognosis for patients aged over 75 years who receive a pacemaker in the context of sinus node dysfunction is unclear.
Aims
We sought to evaluate the incidences of atrial fibrillation, heart failure and death in such patients, and the role of the pacing mode in their prognosis.
Methods
This was a retrospective study of 102 patients aged over 75 years (mean 82.2 ± 4.4 years) who received a pacemaker in the context of sinus node dysfunction.
Results
During the follow-up period (mean 806 days), 36 patients (35.3%) experienced heart failure, 47 patients (46.1%) had an episode of paroxysmal atrial fibrillation, 19 patients (18.6%) progressed to chronic atrial fibrillation and 29 (28.4%) died, the fatal event being sudden death or of cardiac origin in almost half of these patients (44.8%). Patients assigned to dual-chamber minimal ventricular pacing showed significantly lower rates of heart failure episodes ( P = 0.023) and all-cause mortality ( P < 0.001) than those assigned to conventional dual-chamber pacing. In contrast, the two groups did not differ with regard to either paroxysmal or chronic atrial fibrillation.
Conclusion
In patients aged over 75 years, the use of dual-chamber pacemakers incorporating an algorithm minimizing ventricular pacing for sinus node dysfunction seems to decrease the number of heart failure episodes and mortality. On the basis of this finding, the implantation of such devices seems justifiable, even in this age group.
Résumé
Contexte
Le pronostic des patients de plus de 75 ans qui ont bénéficié d’un stimulateur cardiaque pour une dysfonction sinusale est mal connu.
Objectif
Le but de cette étude a été d’évaluer dans cette population, les incidences de la fibrillation auriculaire, de l’insuffisance cardiaque et de la mortalité ainsi que l’influence du mode de stimulation sur le pronostic.
Méthodes
Il s’agit d’une étude rétrospective sur 102 patients de plus de 75 ans porteurs d’un pacemaker pour une dysfonction sinusale.
Résultats
Au cours du suivi (806 jours en moyenne), 36 patients (35,3 %) ont présenté une décompensation cardiaque, 47 patients (46,1 %) un épisode de fibrillation atriale paroxystique, 19 patients (18,6 %) sont passés en fibrillation atriale chronique et 29 (28,4 %) sont décédés dont près de la moitié (44,8 %) de mort subite ou de cause cardiaque. Dans le groupe de patients porteurs d’un stimulateur double chambre avec un algorithme minimisant la stimulation ventriculaire, nous avons observé un taux moindre de décompensations cardiaques ( p = 0,023) et une plus faible mortalité totale ( p < 0,001) que dans le groupe de patients ayant un stimulateur DDD standard. En revanche, il n’y a pas eu de différences entre les deux groupes en termes de fibrillation auriculaire paroxystique ou chronique.
Conclusion
Les stimulateurs cardiaques double chambre avec un algorithme minimisant la stimulation ventriculaire semblent diminuer le nombre d’épisodes de décompensation cardiaque et les décès chez les patients de plus de 75 ans atteints de dysfonction sinusale. Cela justifie donc leur implantation même à cet âge.
Background
Sinus node dysfunction is one of the main indications for pacemaker implantation. However, the prognosis for patients who receive a pacemaker for this reason, in terms of mortality and occurrence of atrial fibrillation (AF) or heart failure (HF), has not been clearly established in the oldest age group.
Large-scale randomized trials, such as MOST and CTOPP , did not really demonstrate a great superiority of dual-chamber pacemakers over single-chamber devices. It has been shown that dual-chamber pacemakers can be deleterious compared with single-chamber ventricular pacemakers when the ventricular stimulation rate is high (MOST , DAVID ). New algorithms were developed in order to limit the ventricular stimulation rate. The SAVE PACe study showed a benefit of dual-chamber minimal ventricular pacing over conventional dual-chamber pacing in terms of persistent AF.
Few patients aged over 75 years were included in these trials. However, the population aged over 75 years corresponds with more than 50% of pacemaker first implantations , and this population is expected to increase threefold within the next 40 years . It therefore seemed worthwhile to analyse the outcome of patients aged over 75 years who had received a pacemaker in the context of sinus node dysfunction, either alone or already associated with paroxystic supraventricular arrhythmia.
Methods
This single-centre, retrospective study, conducted in the Cardiology Department of the Medical, Surgical and Obstetrical Centre in Schiltigheim, France, included all consecutive patients aged over 75 years on the day of pacemaker implantation and referred for initial implantation of a pacemaker in the context of sinus node dysfunction between 1 January 2001 and 31 October 2006. Patients with second- or third-degree atrioventricular block and those who had undergone implantation of a single-chamber ventricular pacemaker were excluded.
The patient’s medical history, clinical characteristics, pacemaker settings and left ventricular ejection fraction (LVEF) at the start of follow-up were extracted from the initial hospitalization report. The clinical events occurring during follow-up were recorded on questionnaires sent to each patient’s cardiologist. The questionnaires were completed with the help of the patient’s primary care physician and the reports from the centre in which the pacemaker had been implanted.
The pacing algorithms considered to be algorithms minimizing ventricular stimulation rate were MVP ® (Medtronic Inc. Minneapolis, MN, USA) and SafeR ® (Sorin Group, Milan, Italy).
The primary endpoints defined before the start of the study were mortality, episodes of HF, incidences of paroxysmal and chronic AF and occurrence of embolic complications. All episodes of HF were taken into account, irrespective of whether they needed hospitalization or only ambulatory treatment adjustment for worsening HF. The causes of death were classified as sudden death, death of cardiac origin or other cause.
Statistical analysis
All analyses were performed according to the “intention-to-treat” principle. Qualitative variables were analysed using the Chi 2 test or Fisher’s exact test when the number of patients concerned was small (<5 for the theoretical population). Student’s t test or a unifactorial analysis of variance was used to compare continuous variables. Data not corresponding to a normal distribution (atrial and ventricular stimulation rates and LVEF) were analysed using a non-parametric test (Mann-Whitney test). Statistical significance was concluded at P < 0.05 (95% confidence interval).
Results
Between 1 January 2001 and 31 October 2006, 102 medical files met all inclusion criteria and these patients were included in the study; their clinical characteristics at the time of pacemaker implantation are summarized in Table 1 . Pacemaker specificities and setting modes are summarized in Table 2 .
| Characteristic | |
|---|---|
| Men | 33 (32.4) |
| Women | 69 (67.6) |
| Age (years) | 82.2 ± 4.4 |
| Sinus node dysfunction without history of AF | 45 (44.1) |
| Sinus node dysfunction with history of paroxysmal AF | 57 (55.9) |
| LVEF at inclusion (%) | 59.3 ± 10.9 |
| Ischaemic cardiopathy | 20 (19.6) |
| Non-obstructive cardiomyopathy | 4 (3.9) |
| Valvular disease | 20 (19.6) |
| Left ventricular hypertrophy | 36 (35.3) |
| Other cardiopathy | 3 (2.9) |
| No known cardiopathy | 37 (36.3) |
| History of hypertension | 80 (78.4) |
| History of heart failure | 16 (15.7) |
| History of stroke | 17 (16.7) |
| Diabetes | 26 (25.5) |
| CHADS 2 score ≥ 2 | 86 (84.3) |
| Signs of heart failure at implantation (Killip class ≥ 2) | 26 (25.5) |
| Pacemaker characteristic/pacing mode | Number of patients (%) |
|---|---|
| Conventional DDD pacing | 71 (69.6) |
| Dual-chamber minimal ventricular pacing | 27 (26.5) |
| AAI pacing | 3 (2.9) |
| VVI pacing | 1 (1.0) |
| Atrial lead positioned into the right atrial appendage | 98 (96.1) |
| Right ventricular lead positioned at the right ventricular apex | 95 (96) |
| Rate-responsive pacing mode activated | 60 (58.9) |
Follow-up
No patient was lost to follow-up. The mean duration of clinical follow-up was 806 days (i.e. approximately 2.2 years), the minimum follow-up being 38 days and the maximum 2043 days.
The pacemaker variables at the end of follow-up are summarized in Table 3 . Events that occurred during the follow-up period are summarized in Table 4 .
| Pacemaker variable | |
|---|---|
| Mean duration of pacemaker follow-up between implantation and last assessment (days) | 670.6 |
| Conventional DDD pacing | 57 (55.9) |
| Dual-chamber minimal ventricular pacing | 28 (27.5) |
| DDI | 2 (2) |
| AAI | 5 (4.9) |
| VDD | 1 (1) |
| VVI | 9 (8.8) |
| Mean (median) percentage of atrial stimulation before any progression to chronic AF | 68.4 (85) |
| Mean (median) percentage of ventricular stimulation before any progression to chronic AF | 56.3 (82.2) |
| Event | |
|---|---|
| Paroxysmal AF not progressing to chronic AF | 34 (33.3) |
| Paroxysmal AF whether or not progressing to chronic AF | 47 (46.1) |
| Chronic AF whether or not preceded by paroxysmal AF | 19 (18.6) |
| Mean (median) time to progression to chronic AF (days) | 535 (391) |
| External electrical cardioversion | 0 (0) |
| Episodes of heart failure | 36 (35.3) |
| Cerebral embolism | 5 (4.9) |
| Other systemic embolism | 1 (1) |
| High-degree atrioventricular block | 14 (13.7) |
| Deaths | 29 (28.4) |
| Sudden death | 5 (17.2) |
| Death of cardiac origin | 8 (27.6) |
| Death from other cause | 16 (55.2) |
| Mean (median) time to death after pacemaker implantation (days) | 598 (557) |
| Complete resolution of symptoms after pacemaker implantation | 74 (72.5) |
| Partial resolution of symptoms after pacemaker implantation | 21 (20.6) |
| No resolution of symptoms after pacemaker implantation | 7 (6.9) |
Fifty-three patients (51.9%) had at least one episode of AF (paroxysmal or chronic) and five patients had no pacemaker interrogation with a programmer during follow-up. During the follow-up period, one patient benefited from a dual-chamber to triple-chamber pacemaker upgrade due to HF.
Among the 38 patients in whom the Holter pacemaker recordings could be determined unequivocally, 24 had their mode of care changed on the basis of these recordings (anticoagulant treatment initiation, antiarrhythmic drug replacement or initiation).
Chronic atrial fibrillation
Patients progressing to chronic AF ( n = 19) had a higher HF incidence ( P = 0.022). Neither the mode of pacing at the time of pacemaker implantation nor the activation of algorithms designed to prevent AF had a significant impact on the incidence of chronic AF. All of these patients were treated with vitamin K antagonists (VKAs) at the end of the follow-up period.
Paroxysmal atrial fibrillation
Patients experiencing one or more episodes of documented paroxysmal AF during follow-up ( n = 47) had more frequently a history of AF before implantation ( P = 0.022) and cardiopathy of any kind of aetiology ( P = 0.037) at the time of pacemaker implantation. A greater proportion of patients in this group experienced one or more episodes of HF ( P = 0.024); they also progressed more frequently to chronic AF ( P = 0.03). In contrast, there were neither more deaths nor more thromboembolic events in this group. The pacing mode did not affect the incidence of paroxysmal AF. With regard to patients who experienced an episode of paroxysmal AF but did not progress to chronic AF, 67.6% received at least transient treatment with VKA during follow-up and 52.9% were still being treated with VKAs at the end of follow-up. None of the treatments ongoing at the time of implantation or during follow-up (in particular angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers) were associated with a statistically significant decrease in the number of paroxysmal AF episodes.
Stroke
Six patients (5.9%) experienced either cerebral or systemic embolism during follow-up. All of these patients had a CHADS 2 score ≥2 . Four of these patients presented documented AF during follow-up, of whom two were not receiving anticoagulant treatment.
Heart failure
Patients experiencing HF during follow-up ( n = 36) had a significantly lower LVEF at inclusion (mean 52.58 vs 62.97%, P < 0.001) and had more frequently ischaemic cardiac disease than another aetiology of cardiopathy ( P = 0.002). These patients were classified more often in Killip class ≥II at the time of pacemaker implantation and were receiving beta-blocker treatment more often at this time. Conventional DDD mode pacemaker programming at the time of implantation was a factor favouring HF ( P = 0.026), whereas dual-chamber minimal ventricular pacing was a factor protecting against this event ( P = 0.033). The ventricular stimulation rate was significantly higher in patients presenting episodes of HF (mean 67.6%; median 96.7%) than in those not experiencing this event (mean 50.6%, median 56.6%) ( P = 0.044). Patients presenting HF episodes during follow-up also had more episodes of paroxysmal AF ( P = 0.024) and chronic AF ( P = 0.022), and had a higher mortality ( P < 0.001).
Mortality
The patients who died during follow-up ( n = 29) were significantly older ( P = 0.033) and had more frequently an LVEF ≤50% ( P = 0.029) than those who survived. The pacemakers of patients who died were more often conventional dual-chamber pacemakers ( P = 0.022) and less often dual-chamber minimal ventricular pacing pacemakers ( P = 0.005). The ventricular stimulation rate was significantly higher among patients who died (mean 78.4%, median 98%) than among those who remained alive (mean 49.4%; median 51%) ( P = 0.039). This group also experienced more episodes of HF ( P < 0.001). In contrast, the rates of paroxysmal and chronic AF did not differ between the two groups.
Comparison of patients assigned to conventional dual-chamber pacing and to dual-chamber minimal ventricular pacing
As the development of pacemakers incorporating an algorithm minimizing ventricular pacing occurred relatively recently, the duration of follow-up of patients who received this type of device was significantly shorter than that of those equipped with a conventional dual-chamber pacemaker (480.6 days vs 919.9 days, P < 0.001). In order to compare these two groups of patients, we perfomed an analysis including only patients with a follow-up duration of less than 1000 days. The main results of this analysis are summarized in Table 5 , and Figs. 1 and 2 .
