Conduction disorders

Apoptosis and fibrofatty infiltration are present during normal ageing of nodal tissue. In addition, in the elderly, sinus node automaticity and conduction disorders may be severely depressed by drugs. In the sinus node, nodal cell rarefaction, perinodal fibrosis and adrenergic receptor damage are partially offset by the decrease in parasympathetic basal tone. In a 75-year-old subject, the number of sinus node cells has declined to 10% compared with in a 20-year-old subject. Membrane channel alterations also impair sinus node automaticity. As a consequence, the intrinsic heart rate, obtained after autonomous blockade, decreases with age. The resting heart rate is not affected, while the exercise heart rate is, leading to reduced exercise capacity. The maximum heart rate decreases by 0.7–1 beat per minute each year in adults. Although the prevalence of sinus node dysfunction is unknown in the elderly, 70–80% of PMs for sinus node dysfunction are implanted over the age of 65 years; after that age, the prevalence of symptomatic sinus bradycardia is estimated at 1/600 .

In the atrioventricular (AV) nodal region, normal ageing causes calcifications of the cardiac fibrous skeleton, which may lead to a slowing of conduction velocity in the AV node, the His bundle or the proximal part of the right and left bundle. Prolongation of the PR interval, which occurs with age, is well known . In elderly subjects, 20% had a PR interval longer than 200 ms in the Manitoba Study , while it exceeded 220 ms in only 4% of the overall population. However, this study showed that the life expectancy of patients presenting with a prolonged PR interval was not impaired during the 30 years of follow-up, compared with subjects with normal AV conduction. In contrast, in an overview of the Framingham study , prolongation of the PR interval was correlated with a higher risk of atrial fibrillation (AF) and global mortality, even when the PR interval remained within the normal range.

Prevalence of type I second-degree AV block was reported to be 2.2% in an observational study involving 625 patients from various age groups (aged 22 to 80 years). Another study on 157 patients reported a PM implantation rate of 61% for type I second-degree AV block. During the 5 years after implantation, quality of life was improved in the youngest patients (aged 45 to 79 years), but not in those aged > 80 years.

Ageing progressively affects the conduction capacity of the terminal part of the intraventricular conduction system . The QRS duration remains stable. The mean frontal QRS axis shifts towards the left, so the prevalence of left axis deviation beyond –30° reaches 20% within the tenth decade. The prognosis value of this axis deviation is unknown. The prevalence of right bundle branch block increases with age, reaching its peak at the age of 70 years in men, but is still increasing at that age in women. In the Baltimore Longitudinal Study of Aging, the mean age at onset of right bundle branch block was 64 ± 14 years. The involvement of right bundle branch block in the onset of heart disease or in mortality seems poor. The incidence and prevalence of left bundle branch block also increases with age; in the Framingham study , its prevalence was 1.1%, the mean age of occurrence was 62 years and, in the majority of the cases, it was associated with a known heart disease or preceded its development. In an Irish cohort study of 110,000 subjects followed between 1968 and 1993, the prevalence of left bundle branch block increased with age in both men and women . AV blocks over type I second-degree affect elderly subjects, with or without heart disease . The prevalence of third-degree AV block remains low. In the Reykjavik study , with a 24-year follow-up among 20,000 individuals, only 11 of them (0.04%) had new-onset third-degree AV block.

It is thus not surprising that the mean age at PM implantation is 75 years, that 80% of the implanted patients are aged > 65 years and that most patients receiving a PM for AV block are aged > 60 years.

Ventricular arrhythmias

As with HF, the incidence of heart rhythm disorders increases with age. In the aging man, the incidence of ventricular arrhythmias increases independent of the presence of underlying heart disease, but few detailed epidemiological data are available .

Cardiac evaluation

Ischaemic, valvular or hypertensive heart diseases are more prevalent with increasing age, and may be complicated by HF during their evolution, which is the main indication for an ICD and/or CRT. Comorbidities, such as AF, peripheral vascular disease, diabetes, pulmonary disease, depression, anaemia and renal dysfunction, are prevalent in one form or another in many elderly patients scheduled for PM or ICD implantation . A search for an extracardiac cause, such as bronchopneumonia, chronic obstructive pulmonary disease, pulmonary embolism or sleep apnoea syndrome, should often be carried out in the context of dyspnoea with suspected HF. The question of whether older patients with multiple comorbidities commonly seen in “real-world” clinical practice will benefit in terms of survival and quality of life from implantation of costly devices is discussed later. These issues are specifically relevant for patients for whom ICD implantation is being considered, and do not really apply to patients needing cardiac pacing or CRT.

The prevalence of HF is between 10% and 20% in those aged 70–80 years , and increases by a factor of 10 between the ages of 60 and 80 years . HF may be the reason (rather than a limitation or non-indication) for device implantation, unless life expectancy is estimated to be dramatically short. Although 50% of HF patients are aged > 75 years, many clinical trials have included younger patients with a mean age of 61 years. Aging HF patients are more often women with less frequent cardiovascular disease and associated risk factors, but with more non-cardiovascular comorbidities, such as chronic renal failure, anaemia or cancer . In aging HF patients, 50% exhibit preserved ejection fraction HF, the long-term prognosis of which is similar to that of HF with depressed ejection fraction . Sudden death occurs 6 to 9 times more frequently in HF patients than in general population. High blood pressure and diabetes are stronger risk factors in aging women, and smoking and coronary heart disease are stronger risk factors in men .

Hypertension may be highly prevalent in PM recipients , but high blood pressure is far less common in patients with severe systolic HF, which is the most common indication for an ICD and/or CRT. Excessively high or low blood pressure may be associated with a poor outcome in patients with coronary artery disease or HF. There may be a reverse association between baseline hypertension (actually protective) and mortality in elderly patients with HF or coronary artery disease . However, blood pressure control in patients with a PM or ICD and hypertension is a relevant goal, in any case, for the prevention of complications, particularly cerebral strokes.

Diabetes commonly coexists, and is a powerful independent predictor of morbidity and mortality, in patients with HF . Diabetes is not a predictor of poor response to CRT , and diabetic HF patients treated with CRT seem to have an outcome that is similar to that of non-diabetic patients . Diabetic patients derive a similar benefit from ICD therapy, despite being possibly sicker and having a higher mortality rate overall .

The prevalence of AF in patients with HF ranges from 5% in patients with New York Heart Association (NYHA) functional class I to 40% in patients with NYHA class IV, and increases markedly with age . AF may be associated with a higher risk of cardiac events and mortality in population studies, but should not per se affect the decision to implant a device at an individual level in the elderly; it may, however, influence the type of the device and its programming. AF patients show significant improvement after CRT, with improvements in left ventricular (LV) ejection fraction that are similar to or slightly greater than those in patients in sinus rhythm . In patients treated with CRT with defibrillation (CRT-D), history of AF is an independent risk factor, not only for mortality, but also for appropriate and inappropriate shocks . Compared with patients with permanent AF, those with paroxysmal or persistent AF may have a lower increase in the risk of mortality or appropriate device therapy, but a higher risk of inappropriate device therapy . Further efforts in AF management may thus optimize the care in elderly patients with electric cardiac devices.

Renal dysfunction is a common comorbidity in HF in the elderly, and is individually associated with poorer outcomes . Renal impairment does not prevent a positive response to CRT, even in elderly patients . CRT might actually be a renal-protective strategy in HF, and improvement in renal function can be another mechanism to explain the beneficial effects of CRT .

Importantly, comorbid conditions, such as renal failure, NYHA stage IV HF, AF or third-degree heart block, are stronger predictors than age in determining higher risk of peri- and post-procedural complications . Elderly patients with comorbidities, but with a good life expectancy, may benefit from pacing, CRT or ICD, although “reasonable expectation of survival” has not been standardized. Koplan et al. found that a proportion of patients aged > 80 years may survive for ≥ 4 years after ICD implantation, despite comorbidities, and survival may reach 6 years for those with an ejection fraction > 30% and a glomerular filtration rate > 60 mL/min, while the overall life expectancy of octogenarians in the USA is 8 years . Goldenberg et al. developed a simple risk model with five clinical factors (NYHA functional class > II, age > 70 years, blood urea nitrogen > 26 mg/dL, QRS duration > 0.12 seconds and AF) for ICD recipients with low ejection fraction. The authors found a pronounced benefit for ICDs in terms of risk of death in intermediate-risk patients, and attenuated efficacy in lower-risk patients (0 risk factors) and higher-risk patients . Thus, comorbidities should certainly be considered when determining whether an elderly patient will markedly benefit from a device or not. Other non-cardiac conditions, such as peripheral vascular disease or pulmonary disease, may have a negative effect on survival in ICD recipients ; this could influence the decision regarding which device to implant (see section entitled “CRT-P or CRT-D?”).

Geriatric assessment

An evaluation before the implantation of an ICD in an elderly person should be accompanied by a geriatric assessment that gauges the person’s overall prognosis, with psychosocial and medical assessments (comorbidities, autonomy). It is necessary to determine the patient’s condition to know whether the person will benefit from the implantation of an ICD in terms of functional capabilities, but also to attempt to predict whether this condition will be maintained over the coming months or semesters or, on the contrary, whether the patient is at risk of deteriorating. The “Comprehensive Geriatric Assessment” (CGA) assesses a person’s state of health, and provides information on functional status, continuing to live at home and overall survival . However, the CGA requires a certain expertise in geriatrics that takes time (mobile geriatric team, day hospital, geriatric care networks) and cannot be offered to all patients ( Table 1 ). Several tools can be used to screen for patients who should undergo a CGA, including the Identification of Seniors at Risk score, the Triage Risk Screening Tool and the G-8.

The French National Authority for Health ( Haute Autorité de santé ) has proposed the presence of the following as factors in decision-making: a “geriatric syndrome” (undernutrition, depression, falls, mental confusion, pressure ulcers); a loss of autonomy existing before hospitalization, based on abnormality in at least one activity of daily living (Activities of Daily Living [ADL] index ; personal hygiene, bathing, dressing, continence, going to the toilet, transferring, feeding; Appendix 1 ); a history of an unplanned hospitalization in the last 6 months; and an unfavourable social situation (precariousness, isolation).

The Haute Autorité de santé has also proposed a score for identifying at-risk individuals, namely the Triage Risk Screening Tool , which can be done in a few minutes by a non-physician and comprises five items: cognitive impairment; difficulty with walking/transferring or recent falls; polymedication (five or more medications per day); history of hospitalization in the last 90 days or admission to an emergency department in the last 30 days; and abnormality before hospitalization in at least one activity of daily living (ADL index) or social isolation.

The cardiology team can detect the main “geriatric syndromes” that should be the subject of a geriatric assessment and more complete treatment if abnormality is detected.

Cognitive impairment

Identifying cognitive impairment can involve the Mini-Mental State Examination or a simpler test, such as the Memory Impairment Screen . The Mini-Mental State Examination has a score range of 30; the threshold value depends on age and sociocultural level. A score of < 27 points hints at possible cognitive impairment; a score of < 24 points hints at possible dementia and warrants a specialized assessment. The Memory Impairment Screen consists of having an individual repeat four words written on a sheet of paper (e.g. CHECKERS, SAUCER, TELEGRAM and RED CROSS) immediately and 10 minutes later. Forgetting one of the words (despite a clue, e.g. “What was the name of the game?”) hints at cognitive impairment and calls for a specialized complementary assessment. Other tests (such as the CODEX tests, the five-word test, the clock test, etc.) can be used to detect cognitive impairment.

Risk of falls

Assessing the risk of falls involves a series of questions (relating to previous falls), a clinical examination (general condition, neuromuscular status, joints, vision and neurological and cardiovascular examinations, looking in particular for orthostatic hypotension) and a few simple tests, such as the stand-on-one-leg test, which assesses an individual’s ability to remain standing on one leg for 5 seconds, and the Timed Up and Go test, which assesses balance and gait (the elderly person must get up from a chair with armrests, walk 3 meters in a straight line to a line drawn on the ground, turn around and sit back down again, with a time of > 20 seconds indicating a risk of falls).

Undernutrition

Undernutrition is defined as weight loss > 5% in 1 month or > 10% in 6 months, and signals the presence of a risk situation. Weight should be interpreted taking into account clinical and biological factors of water retention and dehydration. A nutritional assessment can also be done by determining serum albumin; a value of < 35 g/L indicates undernutrition. Treatment of undernutrition can improve life expectancy.

Depression

The “Mini Geriatric Depression Scale” ( Appendix 2 ) is a quick four-question screening test. If abnormality is detected, the full GDS scale gathers depressive symptoms. A score > 15/30 indicates the possibility of depression, and a score > 22/30 indicates the possibility of major depression. Depression is associated with a more severe cardiovascular prognosis and poorer treatment adherence in the elderly .

Frailty

Detecting frailty (a syndrome of deterioration of physiological reserves and decreased resistance to stress) should also be part of the assessment, as it is associated with unfavourable developments after implantation . The Gérontopôle Frailty Screening Tool in individuals with an ADL score ≥ 5/6 includes six items: living alone; involuntary weight loss in the last 3 months; fatigability in the last 3 months; mobility difficulties in the last 3 months; memory complaints; and slow gait speed (more than 4 seconds to walk 4 meters).

Autonomy

Autonomy can be evaluated by scales that assess activities of daily life (Instrumental ADL and ADL) through questioning the patient and their relatives. The short form of the Instrumental ADL scale includes four items: ability to use the telephone; ability to use transport; responsibility for own medication; and ability to handle finances. The ADL scale gives information about personal hygiene and grooming, dressing, ability to go to the toilet, transferring, continence and ability to eat alone. A subject is considered dependent if they need human help to carry out the activity concerned.

Lifestyle

The assessment should determine the patient’s state of isolation, and take into account the involvement of caregivers (family members or friends) and access to different care services; this includes providing disease information and education to the patient and those close to the patient, to ensure that medications are taken and to attend to complications early.

Thus, the CGA allows a Personalized Care Plan to be prepared before and after intervention, within the framework of a multiprofessional approach ( Fig. 1 ).

Course of care before an implantable cardioverter defibrillator (ICD) after the age of 75 years. ADL: activities of daily living; GFST: Gérontopôle Frailty Screening Tool; TRST: Triage Risk Screening Tool.

Choice among devices with different pacing modes for older patients

The choice of optimal pacing mode for older patients is under debate, as is whether VVI pacing should be preferred for cost reduction (including patients with preserved sinus node function) or whether physiological pacing should be favoured. The choice of pacing mode varies, depending on the disorder to treat (sinus node dysfunction, acquired AV block or permanent AF associated with bradycardia); it also depends on whether bradycardia is paroxysmal or permanent. The 2013 European professional practice recommendations do not specifically discuss age, while the majority of paced patients are now in an older age bracket .

A survival benefit conferred by DDD instead of single-chamber pacing is controversial. DDD pacing may limit the risk of developing AF, stroke or HF. Older studies have compared DDD with AAI pacing, a mode that is rarely used nowadays, having been mostly replaced by algorithms to preserve AV conduction. In a 1997 Danish study, 225 patients (mean age, 71 years) were randomly assigned to VVI versus AAI pacing for sick sinus syndrome, and were followed for up to 8 years . Atrial pacing was associated with longer survival, a shorter duration of AF, less HF and fewer thromboembolic events. In PASE, a study of patients whose mean age was 76 years, the superiority of DDD compared with AAI pacing was mostly confined to patients paced for sinus node dysfunction, and was not significant in cases of AV block . The MOST study compared VVI with DDD pacing in 2010 patients with sinus node dysfunction (mean age, 74 years) . DDD pacing lowered the risk of AF and HF, and improved the quality of life, but did not lower the risk of stroke. The devices had to be reprogrammed from VVI to DDD mode in 20% of patients because of disabling PM syndrome . The CTOPP study enrolled patients with a mean age of 73 years, who presented with a mix of AV block and sinus node dysfunction . Over a follow-up period of more than 6 years, the risk of AF was decreased, but not the risk of death or stroke. In the UK PACE study, 2021 patients aged > 70 years underwent implantation of PMs for AV block, and were randomly assigned to VVI versus DDD pacing . During a follow-up period of up to 5 years, no difference in mortality was observed between the two study groups, and no difference was seen in the development of AF, HF or stroke. In none of these older studies was an algorithm available to preserve AV conduction.

Chao et al. compared 108 PM recipients aged > 90 years with a matched non-paced control group . Cardiac pacing was not associated with a higher mortality, confirming that very old age is not a contraindication to PM implantation. Antonelli et al. studied VVI pacing in nonagenarians over a period of more than 20 years, and observed no adverse effect on survival .

In summary, in older patients with sinus node dysfunction, experts agree that DDD pacing along with the programming of an algorithm to minimize ventricular pacing (see section entitled “Specific algorithms to decrease the percentage of ventricular pacing”) is preferred . Rate responsiveness may be helpful in the elderly, and should be added in cases of sinus node dysfunction associated with chronotropic incompetence. Permanent AV block and no permanent AF should be treated with DDD pacing because of the risk of PM syndrome, which develops in 20% of patients. In very old patients presenting with intermittent AV block or syncope and suspected AV block, VVI pacing may be all that is needed. VVI pacing is also preferred to treat AF associated with bradycardia. These patients may, in the future, be candidates for leadless pacing, broadening the indications for VVIR pacing.

Choice of pacing site

Ventricular pacing cannot be avoided or minimized in patients with permanent complete heart block. Right ventricular (RV) apical pacing is the most common site of lead positioning in the majority of patients receiving PMs, because of the direct access during implantation. Several large randomized clinical trials have demonstrated a direct association between a high percentage of RV apical pacing and a worse clinical outcome; this prompted the development of specific algorithms to decrease unnecessary RV pacing (see section entitled “Specific algorithms to decrease the percentage of ventricular pacing”) and search for alternative pacing sites when pacing is inevitable . Furthermore, apical positioning may be associated with a higher risk of perforation, particularly in old and thin women.

The septum, the RV outflow tract and the His bundle have been suggested as alternative pacing sites to the RV apex. Because of the closer proximity to the normal conduction system, these RV sites may result in less electromechanical dyssynchrony and a more physiological activation. However, whether there is an optimal site for RV pacing remains controversial. A randomized study comparing outflow tract pacing and RV apical pacing failed to show sustained additional clinical benefit . Direct His pacing may, theoretically, be an attractive alternative, but feasibility may be limited . Therefore, the potential benefits of alternative RV pacing sites in old patients remain unclear. The level of proof for systematic implantation of a biventricular device in old patients with complete AV block also remains too limited for this to be proposed in clinical practice.

PM programming in the elderly

Magnetic resonance imaging (MRI) compatibility

After implantation, a substantial proportion of elderly patients with a PM may develop an indication for MRI examination because of medical comorbidities . Implantation of MRI-conditional devices is today becoming a standard of care. Recommendations have been published on how to perform an MRI examination securely in patients with conventional devices or MRI-conditional PM systems . It actually appears that MRI in patients with MRI-conditional pacing systems and selected patients with MRI-unsafe systems can be performed safely under strict conditions in daily clinical practice .