Arrhythmias among Older Adults Receiving Comprehensive Geriatric Care: Prevalence and Associated Factors

Background: Cardiovascular diseases and arrhythmias are medical conditions that increase with age and are associated with significant morbidities and mortality. The aim of the present study was to investigate the prevalence of arrhythmias and clinical associations in the collective of older adults receiving comprehensive geriatric care (CGC). Methods: Holter ECG monitoring (HECG) of older patients hospitalized for CGC was analyzed. The prevalence of arrhythmias and the associations between the presence of arrhythmias, patients’ characteristics and the functional status regarding basic activities of daily living (assessed by the Barthel index (BI)), walking ability (assessed by the timed up and go test (TUG)), and balance and gait (assessed by the Tinetti balance and gait test (TBGT)) were examined. Results: In the presented study, 626 patients were included (mean age: 83.9 ± 6.6 years, 67.7% were female). The most common arrhythmias detected in HECG were premature ventricular contractions (87.2%), premature atrial contractions (71.7%), and atrial fibrillation (22.7%). Atrial flutter was found in 1.0%, paroxysmal supraventricular tachycardia in 5.8%, non-sustained ventricular tachycardia in 12.5%, first-degree AV block in 0.8%, second-degree AV block type Mobitz I in 0.8%, second-degree AV block type Mobitz II in 0.3%, pause > 2.5 s any cause in 3.5%, and pause > 3 s any cause in 1.6% of the cases. Premature atrial contractions were associated with the female sex (74.8% vs. 65.3%, p = 0.018), whereas in male patients, the following arrhythmias were more common: premature ventricular contractions (91.6% vs. 85.1%, p = 0.029), ventricular bigeminus (8.4% vs. 3.8%, p = 0.021), and non-sustained ventricular tachycardia (17.3% vs. 10.1%, p = 0.014). Atrial fibrillation detected in HECG was more frequent in patients at high risk of falls, indicated by their TBGT score ≤ 18 (24.7% vs. 12.0%, p = 0.006), and premature ventricular contractions were more common in patients unable to walk (TUG score 5) compared to those with largely independent mobility (TUG score 1 or 2) (88.0% vs. 75.0%, p = 0.023). In a logistic regression analysis, atrial fibrillation detected in HECG was identified as a risk factor for a high risk of falls (odds ratio (OR): 2.35, 95% confidence interval (CI): 1.23–4.46). Conclusion: In our study, investigation of HECG of older adults hospitalized for CGC revealed that premature atrial contractions, premature ventricular contractions, and atrial fibrillation were the most common arrhythmias. Premature atrial contractions were found to be more frequent in female patients, while male patients were more prone to premature ventricular contractions. In the investigated population, atrial fibrillation emerged as a risk factor associated with a high risk of falls.


Introduction
Cardiac arrhythmias are a prevalent health issue posing a significant challenge for healthcare systems.Age dependency of arrhythmia is observed in several arrhythmia conditions such as atrial and ventricular premature beats, paroxysmal supraventricular tachycardia, non-sustained ventricular tachycardia, atrioventricular block, or particularly, atrial fibrillation (AF) [1,2].An investigation conducted by Khurshid et al. including over 500,000 people, revealed that 2.35% exhibited a baseline rhythm abnormality, with prevalence increasing with age.In this population, AF, bradyarrhythmias, and conduction system diseases were more common than supraventricular and ventricular arrhythmias [3].The presence of cardiac arrhythmia is also relevant in regard to patients' functional status.In a prospective study examining cardiac arrhythmias and physical performance of older people, it was shown that arrhythmia at baseline was associated with the incidence of disability in activities of daily living and worse performance in balance tests [4].AF, as one of the most frequent types of arrhythmia, affected approximately 43.6 million people globally in 2016.The lifetime risk of developing AF for individuals of European ancestry at the age of 55 is one in three [5][6][7].It is projected that the number of people (aged 55 years and over) diagnosed with AF in the European Union will more than double from 8.8 million in 2010 to 17.9 million by 2060 [8].Epidemiological data confirm the highest prevalence of AF among patients 80 years and older with variation from 10 to 17% [9].
The clinical relevance of arrhythmia, especially AF, is significant and evident in its outcomes and complications.It is estimated that 20-30% of all ischemic strokes are caused by AF, and 20-30% of heart failure patients are diagnosed with this condition.AF is associated with increased mortality, cognitive and physical decline, high hospitalization rates, and reduced quality of life [5,10].Frailty, as a syndrome that increases with age causes reduced physical capabilities and due to its relationship to AF, is an important component of current research.Frail patients are more prone to new-onset AF and those with pre-existing AF are at increased risk for serious adverse outcome events such as thromboembolic complications (ischemic stroke), bleeding, and all-cause death [11,12].
The high impact of arrhythmia on health care systems can also be demonstrated by the number of pacemaker implantations, with high-degree atrioventricular block (AV block) and sick sinus syndrome as the most common indications.In the year 2009, approximately 225,000 pacemakers were implanted in the United States and the highest number of new implants was documented in Germany (927 per million).Despite national differences, older adults over the age of 60 consistently represent the largest group of pacemaker recipients [1,13].
Overall, arrhythmia is a commonly diagnosed and diverse clinical condition in everyday clinical practice, and its prevalence is increasing among older adults.For these reasons, we aimed to investigate the diversity and prevalence of arrhythmias in a special population of hospitalized older people who underwent comprehensive geriatric care (CGC).

Patients
From 1231 patients hospitalized in our clinic for geriatrics between May 2019 and April 2020, 626 patients received Holter ECG monitoring (HECG) during their hospital stay and were selected for this retrospective, single-center analysis.

Comprehensive Geriatric Care
Comprehensive geriatric care (CGC) encompasses a specialized treatment framework designed to meet the complex needs of older patients.These patients are referred by their primary care physicians, specialists, or other medical facilities.The scope of treatment within our geriatric department is diverse, encompassing individuals with immediate medical conditions, those in post-operative recovery, and those diagnosed with common geriatric syndromes (such as frailty, mobility issues, dizziness, malnutrition, incontinence, and cognitive decline) [14][15][16].The main objectives of CGC include medical treatment, improvement of functional deficits, and preservation of independence.To achieve these goals, CGC employs a multidisciplinary approach.A thorough assessment is conducted to evaluate mobility, cognitive and emotional capabilities, basic activities of daily living (ADL), and social circumstances.A multiprofessional team, comprising experienced geriatricians, nursing staff, psychologists, social workers, dietitians, and various therapist professions (occupational/physio/speech therapists) collaboratively determine an individualized treatment strategy for each patient.CGC regularly spans a minimum of two weeks, during which a minimum of 20 treatment units (each lasting 30 min) are scheduled.These sessions encompass physiotherapy, occupational therapy, logopedics, or psychological care.The treatment goals for the individual patient are discussed in weekly team conferences and may be adjusted if necessary.

Barthel Index
The Barthel index (BI) is an established assessment tool that serves to evaluate patients' functional independence concerning basic activities of daily living.These encompass essential areas such as dressing, eating, bathing, toileting, and mobility.The BI scale ranges from 0 to 100, with higher scores indicating greater autonomy and capability for independent execution of these activities [17].A BI score of 35 indicates severe motor function impairment and was defined as a cut-off value in our study [18,19].

Tinetti Balance and Gait Test
The Tinetti balance and gait test (TBGT), is a standardized clinical assessment tool to evaluate balance and gait function.This assessment consists of two primary components: the balance assessment, where patients perform tasks such as rising from and sitting down on a chair and attempting to rise and turning 360 • , and the gait assessment, which involves observing aspects of walking like step pattern and trunk stability.The total score achievable in the TBGT is 28 points.The higher the score, the better the performance [20].A TBGT score ≤18 implicates high risk of falling and was selected as a cut-off value in the presented investigation [21,22].

Timed up and Go Test (TUG)
The timed up and go test (TUG) is a standardized clinical assessment to evaluate patients' walking ability and potential fall risk.In this test, the patient sits on a chair, stands up, walks three meters, turns around, returns to the chair, and sits down.The time taken to complete the test is measured, with an extended duration indicating compromised mobility or an elevated likelihood of falling.Defined TUG categories: (5) patient not able to walk; (4) >30 s performing TUG; (3) 20-29 s performing the TUG; (2) 10-19 s for TUG completion; (1) <10 s for TUG completion.A TUG score of 1 indicates unrestricted walking ability, patients with a TUG score of 2 tend to be independently mobile, and a TUG score of 3 or 4 implicates mobility restriction relevant to daily life [23].

Holter ECG Monitoring
Holter ECG monitoring (HECG) is a routinely performed diagnostic procedure in our geriatric department.HECG was scheduled for 24 h with wearable ECG devices (custo flash 510 and custo flash 200, custo med GmbH, Ottobrunn, Germany).
Data sets were stored and evaluated offline within the manufacturer's provided system (custo diagnostic, custo med GmbH, Ottobrunn, Germany).Examinations were evaluated by a cardiologist in the clinical care routine.For this retrospective study, findings of the routinely performed HECG examinations were evaluated for the presence of the following arrhythmias: premature atrial contractions (PACs), premature ventricular contractions (PVCs), ventricular bigeminus, ventricular couplet and non-sustained ventricular tachycardia (NSVT, defined as run of consecutive ventricular beats persisting for 3 beats to 30 s [24]), paroxysmal supraventricular tachycardia (PSVT), atrial fibrillation (AF), atrial flutter, first to third degree atrioventricular block (AV block), pause > 2.5 s, and pause > 3 s.
Indication for HECG was based on individual physicians' judgement.Reviewing the case files, the following disorders/reasons for performing HECG were identified:

Data Collection
For the present study, the case files of patients who were treated in our geriatric department between May 2019 and April 2020 and received HECG during their hospital stay were evaluated (n = 626).Demographic parameters (age and sex), patients' comorbidities (listed in Table 1; acute and previous stroke/intracranial hemorrhage, intracranial tumor, and unspecified head injuries were summarized as structural brain lesion), HECG findings (illustrated in Figure 1), and scores in the geriatric assessment (BI, TBGT, and TUG) on hospital admission were documented for further statistical analysis.

Statistical Analyses
Demographic parameters, comorbidities, and HECG findings were documented in the clinical care routine and evaluated for this retrospective analysis.Data sets were presented as mean ± standard deviation (SD), median, and interquartile range (IQR), as well as percentages and frequencies.Fisher's exact test for categorical data was carried out.Normal distribution was checked by the Kolmogorov-Smirnov test.Analysis of unpaired samples was carried out using the Mann-Whitney U-test for non-normally distributed data and the t-test in case of normal distribution.Clinical factors associated with high risk of falls indicated by a TBGT score ≤ 18 (with a p-value < 0.05 between groups in univariate analysis) were entered into logistic regression analysis.Statistical analysis was performed using PSPP software (version 1.4.1,GNU project).

Ethical Approval
We received ethical approval for this retrospective study (Ethical committee of the Medical Chamber Westfalen-Lippe and of the Westfälische Wilhelms-Universität Münster (University of Münster, Germany) protocol number: 2021-175-f-S).

Statistical Analyses
Demographic parameters, comorbidities, and HECG findings were documented in the clinical care routine and evaluated for this retrospective analysis.Data sets were presented as mean ± standard deviation (SD), median, and interquartile range (IQR), as well as percentages and frequencies.Fisher's exact test for categorical data was carried out.Normal distribution was checked by the Kolmogorov-Smirnov test.Analysis of unpaired samples was carried out using the Mann-Whitney U-test for non-normally distributed data and the t-test in case of normal distribution.Clinical factors associated with high risk of falls indicated by a TBGT score ≤ 18 (with a p-value < 0.05 between groups in univariate analysis) were entered into logistic regression analysis.Statistical analysis was performed using PSPP software (version 1.4.1,GNU project).

Ethical Approval
We received ethical approval for this retrospective study (Ethical committee of the Medical Chamber Westfalen-Lippe and of the Westfälische Wilhelms-Universität Münster (University of Münster, Germany) protocol number: 2021-175-f-S).
Patients' characteristics by performance in functional assessments are summarized in Tables 2-4.

Discussion
In the present study, the highest arrhythmia prevalence was documented for ectopic beats with 71.7% for premature atrial contractions and 87.2% for premature ventricular contractions, respectively.Atrial fibrillation as a high-risk factor for disabling diseases such as stroke and heart failure was found in 22.7% of the HECG.
Arrhythmia is a clinical condition that increases with age based on a variety of pathophysiological processes.Aging is associated with progressive degenerative changes in the contractile and conduction systems.Electrical remodeling, structural changes, autonomic nervous system dysregulation, electrolyte imbalances, oxidative stress, concomitant diseases, and medication side effects disrupt the heart's normal electrical activity and may contribute to the development of supraventricular and ventricular arrhythmias [1,[25][26][27].
Considering frequencies of arrhythmias in our study, PVCs and PACs emerged as the most common arrhythmia patterns in the HECG examinations.This finding aligns with several previous investigations [2,28,29].The results of the cardiovascular health study by Maniolo et al., investigating arrhythmia in 24 h ambulatory ECG of older adults, revealed that both ventricular and supraventricular ectopic activity are extremely common in older people, with only 18% and 2.8%, respectively, showing no such occurrences [29].In a separate study focusing on PACs in Japanese men, it was shown that 99% of participants had at least one PAC during 1 h.The authors found a significant increase in the number of PACs with age, with the median number of PACs per hour rising from 1.04 in patients aged 40-59 to 4.19 in those aged 70-79 years old [2].Conen et al. investigated PACs in the general population and reached similar conclusions on PAC frequency and age distribution [30].
Regarding PVCs, a high prevalence has also been supported by a series of examinations [28,[31][32][33].In a recent study from Dong et al., which focused on outpatients with palpitations, the prevalence of PVCs was found to be 67.7%,increasing with age and reaching 84.6% in individuals aged 80 years and older.In the study population of Dong and colleagues, a significantly higher PVC prevalence was observed in men, with the greatest sex-specific differences seen in individuals over 60 years of age [28].These findings are consistent with the results of our examination, that also found a higher frequency of PVCs in men.Overall, data on sex distribution of PVCs are not clear.In contrast to our findings and those of Dong et al., a population-based cohort study found a higher age-adjusted incidence of symptomatic PVCs in females [34].The reasons for these sex-specific frequency differences are speculative.It could be discussed that the study populations are quite different.
Sirichand and colleagues investigated a cohort with an average age of 52.1 years, while our study population had a mean age of 83.9 years, more than 1.5 times higher, and had higher rates of cardiovascular risk factors [34].In regard to patients' functionality, performance of less physical activity was associated with greater PVC frequency, as indicated by the investigation of Marcus et al. [35].This assertion is further substantiated by the findings of our investigation, which revealed that PVCs were particularly prevalent in patients unable to walk compared to those with largely unrestricted mobility.In the present study, the risk factors diabetes mellitus and coronary heart disease, as arrhythmia-favoring conditions, were significantly more common in men than in women.Overall, the clinical relevance of ectopic beats can be discussed controversially, since it is also a very common phenomenon in the general population.However, few studies have shown that PVCs have prognostic relevance in regard to mortality of people with and without heart disease [31,32].
Considering NSVT, an extensive investigation by Lin et al. revealed its prognostic significance in relation to various outcomes, including death, cardiovascular-related hospitalization, ischemic stroke, and new-onset heart failure.In this population, patients with NSVT were older and predominantly male; diabetes mellitus and hypertension diagnoses were associated with NSVT occurrence.The findings of Lin et al. support the results of the present study, which also identified that males are particularly affected by NSVT [36].
In our investigation, AF is the second most common arrhythmia after ectopic beats and was observed in 22.7% of the HECG.AF poses a significant challenge for healthcare systems in the 21st century, given its escalating prevalence and potential for severe associated conditions, such as stroke and heart failure.In a cross-sectional examination conducted in Italy among subjects aged 65 and older, the prevalence of AF was estimated to be approximately 1.08 million in 2016, and it is projected to increase by 75% to about 1.89 million people by 2060.Similar predictions have been made by the authors for the European Union, with an estimated 89% increase in AF prevalence from 2016 to 2060.In 2016, 53.5% of individuals diagnosed with AF in the Italian study population were aged at least 80 years, and it is projected that this percentage will rise to 69.6% in that age group by 2060.In the European Union, AF prevalence is predicted to increase from 51.2% to 65.2% during the same time period among older adults 80 years and above [37].Previous data indicate a higher prevalence of AF in men [38,39].When examining sex differences in cardiovascular comorbidities within our study population, it becomes evident that the diagnosis of AF is also more frequent in men.However, this sex discrepancy is not significantly pronounced in the HECG findings.The rate of detectable AF during the inpatient stay, as recorded in the HECG, was notably lower than the proportion of patients diagnosed with AF (22.7% versus 38.5%).This point refers, on the one hand, to the potential lack of detectability in cases of paroxysmal AF during HECG, and on the other hand, to the effects of rhythm-controlling measures.
AF is not only clinically significant due to its high prevalence, but it also holds considerable prognostic importance.The presence of AF impacts both mortality and morbidity in affected individuals, leads to restrictions of the functional status, and furthermore impacts quality of life [5].The impact of AF on quality of life and exercise tolerance is evident in more than 60% of AF patients, with women, younger individuals, and those with comorbidities particularly affected [5,[40][41][42][43].AF prevalence in relation to patients' functionality is of particular relevance.Several studies revealed the association between AF and functional impairment.Frailty as a syndrome of reduced physical capabilities and enhanced vulnerability is a common phenomenon among older adults diagnosed with AF.On the one hand, frail patients are at higher risk for new-onset AF and on the other hand, frail patients already diagnosed with AF are at higher risk for all-cause death, ischemic stroke, and bleeding [11,12].In addition to that, frail patients were less likely to be treated with oral anticoagulants, which might increase the risk of further thromboembolic complications [44].Considering patients' functional status, we found that AF detected in HECG during patients' hospital stay was associated with a high risk of falls, indicated by TBGT score ≤ 18. Supporting our results, Koca and colleagues documented higher fall rates among AF patients in their study [10].These results align with the research conducted by Hung et al. who also demonstrated that AF diagnosis is associated with falls in the older population [45].Furthermore, disability in basic and instrumental activities of daily living and slower walking speed was observed in patients with an AF diagnosis [10,46].In an examination by Donoghue et al. focusing on community-dwelling older adults, associations between AF and mobility were investigated.The study revealed that AF was independently linked to lower usual gait speed, and this effect was more pronounced in participants aged 70 years and older [47].
The frequency of other cardiac arrhythmias, which were found in addition to AF and ectopic beats in HECG, was much lower.Nevertheless, the occurrence of these arrhythmias is not less relevant in clinical practice.Thus, for example, higher-grade AV block is particularly significant for those affected, as these cardiac arrhythmias may require the use of a pacemaker device [1].Higher-grade AV block second-degree type Mobitz II was found in 0.3% of the HECG, and third-degree AV block was not detected.For comparison, a Chinese study with a very large sample size of 15 million participants revealed prevalence rates of 0.42‰ for second-degree AV block, and 0.1‰ for third-degree AV block in participants aged 60 years and older [48].
As with further arrhythmias like AF, the occurrence of AV block was associated with age and cardiac risk factors such as hypertension and diabetes mellitus, respectively [48][49][50].Contrary to the findings of other studies, our research did not reveal a sex-specific association with a more frequent occurrence of AV block in males [48][49][50].
In summary, our study offers an overview of the prevalence of various cardiac arrhythmias in older adults, providing valuable insights into this age group's cardiac health.
The major strength of this study is its substantial sample size, consisting of 626 hospitalized geriatric patients, with a notably high mean age of 83.9 years.Furthermore, the study benefits from the inclusion of a diverse treatment spectrum offered in the geriatric clinic, encompassing a wide range of surgical and medical diagnoses.This allows for a comprehensive analysis of a broad patient population.However, limitations should also be taken into account.A major limitation of the present study is the absence of a control group.To address this limitation and provide valuable insights into the epidemiology of arrhythmia in older people, future research projects could consider comparing the prevalence of arrhythmia between older adults requiring inpatient treatment and community-dwelling, healthy individuals within the same age range.A selection bias could be considered as a further limitation of the study is that it was an individual geriatrician's decision which patients were selected for CGC and underwent HECG.

Conclusions
In our study investigating HECG of older adults hospitalized for CGC, premature atrial contractions, premature ventricular contractions, and atrial fibrillation were the most common arrhythmias.Premature atrial contractions were found to be more frequent in female patients, while male patients were more prone to premature ventricular contractions.In the investigated population, atrial fibrillation emerged as a risk factor associated with a high risk of falls.
# includes acute and previous stroke/intracranial hemorrhage, intracranial tumor, and unspecified head injuries.* presented as median and interquartile range (IQR) and mean ± standard deviation (SD).

Table 3 .
Comorbidities and Holter ECG findings by patients' score in timed up and go test (TUG).Patients unable to walk vs. patients with largely unrestricted walking ability (TUG 5 vs. TUG 1 or 2) and vs. those with mobility restrictions (TUG 5 vs. TUG 3 or 4).
# includes acute and previous stroke/intracranial hemorrhage, intracranial tumor, and unspecified head injuries.* presented as median and interquartile range (IQR) and mean ± standard deviation (SD).

Table 4 .
Comorbidities and Holter ECG findings by Tinetti balance and gait test (TBGT).TBGT ≤ 18 indicates high risk of falling.
# includes acute and previous stroke/intracranial hemorrhage, intracranial tumor, and unspecified head injuries.* presented as median and interquartile range (IQR) and mean ± standard deviation (SD).

Table 5 .
Comorbidities and Holter ECG findings by sex.
# includes acute and previous stroke/intracranial hemorrhage, intracranial tumor, and unspecified head injuries.* presented as median and interquartile range (IQR) and mean ± standard deviation (SD).

Table 6 .
Logistic regression analysis: clinical factors associated with high risk of falls (indicated by Tinetti balance and gait test (TBGT) score ≤18).