Abstract
Background
Frailty is a geriatric syndrome that leads to impairment in interrelated physiological systems and progressive homeostatic dysregulation in physiological systems.
Objective
The focus of the present systematic review was to study the association between the activity of the cardiac autonomic nervous system (ANS) and frailty.
Methods
A systematic literature search was conducted in multiple databases: PubMed/MEDLINE, Embase, Cochrane Library, Web of Science, CINAHL and ClinicalTrials.gov; the last search was performed in March 2015. Inclusion criteria included: 1) the studied population was classified for frailty according to a standard definition, such as the Fried’s criteria; 2) had a non-frail control group; 3) heart rate (HR) and/or heart rate variability (HRV) were parameters of interest in the study.
Results
Of the 1544 articles screened, 54 were selected for full text review and six studies met inclusion criteria. Assessment of HRV using different standard time-domain, frequency-domain, and non-linear domain approaches confirmed the presence of an impaired cardiac ANS function in frail compared to non-frail participants. Furthermore, HR changes while performing a clinical test (e.g., seated step and lying to standing tests) were decreased in the frail group compared to the non-frail group.
Conclusions
The current systematic review provides evidence that the cardiac ANS is impaired in frail, compared to non-frail, older adults as indicated by a reduction in the complexity of HR dynamics, reduced HRV, and reduced HR changes in response to daily activities. Four out of six included articles recruited only female participants and in the other two articles the effect of gender on impairment of cardiac ANS was insufficiently investigated. Therefore, further studies are required to study the association between cardiac ANS impairments and frailty in males. Furthermore, HRV was studied only during static postures such as sitting, or without considering the level of activity as a potential confounder. Accordingly, simultaneous measurement of both physiological (i.e., HRV) and kinematic (e.g., using wearable sensor technology) information, may provide a better understanding of cardiac ANS impairments with frailty while controlling for activity.
Keywords: Autonomic Nervous System, Sympathovagal balance, Sympathetic, Parasympathetic, Heart Rate, Heart Rate Variability, Frailty, Older Adults, Geriatric, Aging
Introduction
Frailty is a geriatric syndrome defined as impairment in interrelated physiological systems, coupled with a decrease in physiological reserves [1, 2], which results in an increased vulnerability to stressors. Accordingly, frailty is associated with adverse health outcomes including mortality, morbidity, postoperative complications, institutionalization, and hospitalization [2–4]. Frail individuals are more susceptible to both acute and chronic conditions such as myocardial infarction, heart disease, and cardiac surgery [5, 6]. Frailty is higher in older adults with cardiovascular disease with a prevalence range from 10% to 60% depending on population studied and the frailty assessment approach [7, 8]. Also, results of a recent study highlighted that early stages frailty (i.e., pre-frailty) contributes to development of cardiovascular disease [9]. Assessment of frailty and its impact on cardiovascular systems is, therefore, important in older vulnerable population.
Frail elders are homeostenotic, and have decreased ability to maintain homeostasis under stress. Autonomic nervous system (ANS) including cardiac ANS and their balanced functioning have a crucial role in maintaining homeostasis in almost all physiological functions [10]. This role becomes more pronounced during responding to internal or external stressors such as disease and activities of daily living [10, 11]. It has been postulated that frail older adults exhibit a progressive homeostatic dysregulation in physiological systems, including the cardiac ANS [9]. Also, impairment in cardiac ANS as highlighted in previous research studies [12, 13] may be a factor that accelerates the frailty process. For example, frailty and heart failure are frequently associated together, with a high prevalence of frailty in heart failure ranging from 15% to 74% [14]. It is known that both heart failure and its treatment can result in, or worsen frailty [15]. Impact of heart failure on frailty status may happen due to changes in the cardiac ANS activity (e.g. increase sympathetic activity) in order to restore cardiac output [16, 17], which in return may add overload on cardiac ANS system that already suffers from a decreased physiological reserves. Also, the impairment of cardiac ANS due to frailty may further worsen heart failure because it is not able to maintain the hemodynamic homeostasis.
As an aid to facilitate understanding of the cardiac ANS mechanism in association with frailty status, non-invasive objective assessment methods based on heart rate (HR) and heart rate variability (HRV) have been introduced [18, 19]. The degradation of cardiac autonomic control results in loss of complexity in the HR and a reduction in HRV, which is associated with a higher morbidity and mortality in individuals with myocardial infarction and heart failure [20, 21]. Cardiac ANS parameters associated with frailty, such as HR and HRV can capture the reduction in dynamics of physiological system and in order to distinguish between healthy and impaired heart functioning [19]. Also, HR and HRV parameters could be used to enhance the prediction of hospitalization outcomes and mortality, especially in older adults with heart disease, as well as to enhance surgical risk assessment and assess the benefits of interventions to prevent frailty or reverse frailty process.
The objectives of this Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) literature review were to: 1) investigate protocols used for cardiac ANS assessment in individuals with different frailty levels defined by a standardized frailty criteria; and 2) to study the relationship of cardiac ANS, as manifested by HR derived parameters (HR and HRV), and frailty status. Findings of the current review article would provide insight regarding cardiac ANS alterations with frailty that would assist with evaluation and management of cardiac conditions in older adults. Also, outcomes of current article would highlight if the measures of cardiac homeostatic integrity can be used as a complementary tool along with traditional approaches to improve frailty assessment.
Methods
Inclusion/exclusion Criteria
A systematic literature review was performed based on PRISMA guidelines for systematic review [22] (see Appendix A for PRISMA checklist).
Inclusion/exclusion criteria were chosen using a focused interview method by experts in the field of aging and frailty, cardiology, clinical research, library science, and biomedical engineering. Inclusion criteria include: 1) a frail sample, as explicitly defined according to a standard definition of frailty such as the Fried’s criteria [2] was studied; 2) a control group (non-frail) was considered; and 3) HR and/or HRV were considered as parameters of interest. The following were considered as exclusion criteria: case reports, letters, systematic reviews/meta-analyses, data published in the form of abstracts instead of peer-reviewed articles, non-English studies, and animal studies.
Search Strategy
Search terms, including controlled vocabulary terms (e.g., MeSH) and key words, were compiled and iteratively refined by content experts in the fields of library science, geriatrics, and biomedical engineering. A medical librarian (CH) then constructed and performed the searches in the following databases: PubMed/MEDLINE (1946–2015), Embase (1947–2015), Cochrane Library (1898–2015), Web of Science (1898–2015), EBSCO/CINAHL (Plus with Full Text) (1981–2015), and ClinicalTrials.gov (1997–2015). The last literature searches were completed in March 2015. The complete PubMed/MEDLINE search strategy, upon which the other database searches were also built, is available in Appendix B. Citations to and reference lists within the selected articles were also searched for studies that would meet inclusion criteria. The World Wide Web (especially Google Scholar) was also perused for relevant articles.
Study Selection
Two independent reviewers performed the study selection (SP, BH). In case of disagreements, a third reviewer (JM) cast the deciding vote. Screening was done in two steps (Figure 1). Initially, titles and abstracts of retrieved references were screened for inclusion in full paper review. Then, the full texts of articles thus selected were further analysed to see if they fulfil inclusion and exclusion criteria.
Figure 1.
Flowchart of the process of literature search and extraction of studies meeting the inclusion criteria
Data Extraction
Two individuals (SP, BH) independently extracted the following data from the included articles: study goals; subjects and study design; definition of frailty used; protocol and equipment used to monitor HR and rhythm; and HR and HRV parameters (and their definitions). Any disagreements were resolved by consensus (Tables 1, 3, and 4).
Table 1.
Study characteristics and research goals with respect to assessment of association of cardiac autonomic nervous system and frailty
| Reference | Study goals | Subjects & study design | Frailty definition | Protocol for heart rate monitoring | Heart monitoring instrument | Comorbidity and Medication Considerations |
|---|---|---|---|---|---|---|
| Chaves et al. 2008 | To study the physiological complexity underlying regulation of heart rate dynamics and its association with frailty |
N: 389 women Non-frail: 259 Frail: 130 Age: 65–95 Participants: Moderate to severe physical disability Setting: Women’s Health and Aging Study I, Community dwelling Design: Cross sectional, Retro-respective |
Fried criteria [2] | Assessment during physical tasks, including walking, isometric exercise, grip strength with diverse positions (lying, sitting, standing) | Two channel holter monitoring -SpaceLabs (model 90205) |
Adjusted their model based on the following medications and comorbidities: -Medication: beta-blockers -Comorbidities: coronary heart disease, congestive heart failure, diabetes |
| Varadhan et al. 2009 | To define a new measure of heart rate variability to capture the impairment in cardiac autonomic control due to frailty |
N: 276 women Age: 65–101 Participants: Moderate to severe physical disability Setting: Women’s Health and Aging Study I, Community dwelling Design: Cross sectional, Retro-respective |
Fried criteria [2] | Assessment during physical tasks, including walking, isometric exercise, grip strength with diverse positions (lying, sitting, standing) | Two channel holter monitoring -SpaceLabs (model 90205) |
Adjusted their model based on the following medications and comorbidities: -Medication: beta-blockers -Comorbidities: cardiovascular disease, diabetes |
| Weiss et al. 2009 | 1) To assess the relationship between physiological systems (cardiac, pulmonary and musculoskeletal) and exercise capacity 2) To assess the effect of seated step test on exercise capacity in different frailty status |
N: 547 women Non-frail: 416 Frail: 131 Age: >=65 Participants: Moderate to severe physical disability Setting: Women’s Health and Aging Study I, Community dwelling Design: Cross sectional, Retro-respective |
Fried criteria [2] | Assessment during seated step test: performing a step like motion while seated in a chair with 4 different height of step in each stages (each stage is 3 min) | Not mentioned | Medications (pulse-blocking medication) and comorbidities (coronary artery disease, chronic lung disease) were not significantly different between non-frail and frail group. |
| Romero-Ortuno et al. 2011 | To classify a community sample of older people into three increasing frailty phenotypes and compare their orthostatic hemodynamic |
N: 442 (317 women) Non-frail: 198 (Age: 70.5±6.3, 66.7% women) Pre-frail: 213 (Age: 72.9±7.5, 76.1% women) Frail: 31 (Age: 76.6±7.3, 74.2% women) Age: >=60 (Mean age: 72.1) Participants: Non-frail, Pre-frail, Frail Setting: Geriatric research clinic, Community dwelling Design: Cross-sectional |
Modified Fried criteria [16] | Assessment during Lying-to-standing orthostatic test | Heart rate monitoring -Finapres Medical Systems BV (Finometer Pro device) |
Subjects with following medications and comorbidities were excluded from study: -Medication: adrenergic antagonists -Comorbidities: ventricular premature complexes, sinus or supraventricular tachycardia, atrioventricular blocks, and atrial fibrillation |
| Takahashi et al. 2014 | To evaluate the short-term complexity of heart rate dynamics in resting supine and standing positions and its association with frailty |
N: 80 (56 women) Non-frail: 38 (Age: 67.1±6.2, 76% women) Pre-frail: 36 (Age: 70.4±7.5, 64% women) Frail: 6 (Age: 86.0±5.8, 67% women) Age: 60–94 Participants: Non-frail, Pre-frail, Frail Setting: Community dwelling Design: Cross-sectional |
Fried criteria [2] | Assessment during 10 minutes in the supine position and 10 minutes in the standing position. During the monitoring, participants asked to speak only when necessary and to breathe spontaneously. | 12-leads ECG monitoring | Subjects with following comorbidities were excluded from study: -Medication: adrenergic antagonists -Comorbidities: diabetes, Parkinson, severe chronic renal failure, vitamin B12 or red cell folate deficiency |
| Katayama et al. 2015 | To study association of cardiac autonomic modulation and frailty status |
N: 23 women Non-frail: 8 (Age: 70.0±2.2) Pre-frail: 8 (Age: 67.0±2.1) Frail: 7 (Age: 73.0±3.5) Age: 65–77 Participants: Non-frail, Pre-frail, Frail Setting: Community dwelling Design: Cross-sectional |
Fried criteria [2] | Assessment during 10 minutes in the supine position with spontaneously breathing and without moving and talking. | Heart rate monitoring -Polar Electro (RS800 CX) |
Subjects with following medications and comorbidities were excluded from study: -Medication: beta-blockers, antiarrhythmic, tricyclic antidepressants and centrally acting antihypertensive agents and/or hormone replacement therapy -Comorbidities: neurological disease, Chagas disease, having cardiovascular events during last one year, and using cardiac pacemaker |
Table 3.
Heart rate and heart rate variability (HRV) parameters reported in studies included in this review
| Reference | Assessment | Monitoring duration | HR indices | HRV indices |
|---|---|---|---|---|
| Chaves et al. 2008 | HRV | 2–3 hours | – | - Time domain: SDNN, SDANN,RMSSD - Frequency domain: TP, pVLF, pLF, pHF and LF/HF - Non-linear domain: ApEn HR |
| Varadhan et al. 2009 | HRV | 2–3 hours | – | - Time domain: SDNN, RMSSD, pNN50 - Frequency domain:, pVLF, pLF, pHF and LF/HF - PC1 and PC2: Principal component of HRV measures |
| Weiss et al. 2009 | HR | During the test | - Baseline HR, [beats/min] - Delta HR, [beats/min]: Change in HR beginning to end |
– |
| Romero-Ortuno et al. 2011 | HR | From beginning of the test until 3 minutes after standing | - Baseline HR, [beats/min] - Delta HR, [beats/min]: maximum HR within 30s-baseline HR - Maximum HR by 30s, [as % of baseline HR] - HR by 120s, [as % of baseline HR] |
– |
| Takahashi et al. 2014 | HRV | Stationary sequence of 256 heart beats during each condition (supine and standing) | – | - Non-linear domain: ApEnHR,CEHR |
| Katayama et al. 2015 | HR HRV |
Stationary 5-min period | - Mean RR, [ms] | - Time domain: SDNN, RMSSD - Frequency domain: pLF, pHF and LF/HF - Non-linear domain: ApEnHR, SEHR |
SDNN = Standard deviation of all Normal to Normal (N-N) intervals; SDANN = Standard deviation of the average N-N intervals over 5-minute periods; RMSSD = Square root of the mean squared differences of successive N-N intervals; pNN50 = Proportion of all N-N intervals that are larger than 50 ms; TP = power across all frequency bands (<0.4Hz); pVLF = power density in very low frequency band (<0.04 Hz); pLF = power density in low frequency band (0.04–0.15 Hz); pHF = power density in high frequency band (0.15–0.4 Hz); LF/HF= ratio between power in low frequency band to power in high frequency band; PC1 = The mean of log-transformed HRV indices; PC2 = Linear combination of log-transformed indices; ApEnHR = Approximate entropy for a heart rate time series; CEHR = Conditional entropy for a heart rate time series; SEHR = Sample entropy for a heart rate time series; RR is the corresponding to the time between consecutive R-waves in ECG and has a reciprocal relation with heart rate
Table 4.
Mean values of heart rate and heart rate variability parameters for different frailty categories. Summary of findings are presented for studies that do not provide quantitative values.
| Reference | Parameters and Results | ||||
|---|---|---|---|---|---|
| Chaves et al. 2008 | - Median ApEnHR was lower in frail compared to non-frail (P-value=0.03). - Analysis of HRV revealed that a lower SDNN, SDANN, TP, pVLF, pLF and LF/HF were associated with a higher probability of frailty (P-value<0.05). However, no significant association was observed for RMSSD and pHF. |
||||
|
| |||||
| Varadhan et al. 2009 | - Second principal component (PC2) used for aggregation of recorded traditional HRV measures was the best predictor of Frailty (P-value<10−5). - PC2 can capture the long-range correlation in heartbeat fluctuations across frequency bands (VLF to HF). - pVLF, pLF, SDNN, LF/HF and PC2 had a significant and negative association with frailty (P-value<0.05). |
||||
|
| |||||
| Weiss et al. 2009 | Non-frail | Frail | P-value | ||
| Baseline HR, [beats/min] | 74 | 75 | 0.486, n.s. | ||
| Delta HR, [beats/min] | 16 | 13 | 0.001*, ↓ in frail | ||
|
| |||||
| Romero-Ortuno et al. 2011 | Non-frail | Pre-frail | Frail | P-value | |
| Baseline HR,[beats/min] | 67.7±10.7 | 68.8±10.6 | 73.0±9.2 | 0.008*, ↑ in frail | |
| Delta HR, [beats/min] | 15.5±6.2 | 13.7±10.5 | 13.3±7.1 | <0.001*, ↓ in frail | |
| Maximum HR by 30s, [as % of baseline HR] | 123.7±10.9 | 120.6±18.0 | 118.8±11.2 | <0.001*, ↓ in frail | |
| HR by 120s, [as % of baseline HR] | 110.1±10.5 | 109.0±10.8 | 111.8±28.3 | 0.675, n.s. | |
|
| |||||
| Takahashi et al. 2014 | Non-frail | Pre-frail | Frail | P-value | |
| CI (supine) | 0.73±0.08 | 0.74±0.08 | 0.79±0.09 | >0.05, n.s | |
| CI (stand) | 0.67±0.08 | 0.69±0.12 | 0.73±0.08 | >0.05, n.s | |
| ApEnHR (supine) | 3.90±0.48 | 3.86±0.46 | 4.07±0.33 | <0.05*, ↑ in frail | |
| ApEnHR (stand) | 3.55±0.58 | 3.58±0.67 | 3.93±0.46 | <0.05, ↑ in frail | |
|
| |||||
| Katayama et al. 2015 | - Non-significant higher mean RR (lower mean HR) observed in non-frail compared to pre-frail and frail in supine position. - A non-significant trend to lower HRV captured by SDNN and RMSSD in frail group compared to pre-frail and frail. Reduction in SDNN and RMSSD in frail group can be an indicator of reduced global HRV and decrease in parasympathetic activity, respectively. - Frail group showed a significant decrease in pHF compared to non-frail group (P-value<0.05) which is an indicator of reduction in activity of parasympathetic nervous system. - Frail group presented significant increase in pLF and LF/HF as compared to the other groups (P-value<0.05). - Regarding non-linear methods, symbolic analysis revealed that frail group had a significant lower incidence of patterns in 2LV family compared to non-frail group (P-value<0.05). This suggests that parasympathetic activity is decreased in frail as compared to non-frail. - A non-significant trend towards a lower SE in pre-frail and frail groups compared to non-frail group was observed. It suggests that frailty is linked to lower physiological complexity. |
||||
↑ = Significant increase; ↓ = Significant decrease; n.s. = Not significant;
= Statistically Significant
HRV = Heart Rate Variability; SDNN = Standard deviation of all Normal to Normal (N-N) intervals; SDANN = Standard deviation of the average N-N intervals over 5-minute periods; RMSSD = Square root of the mean squared differences of successive N-N intervals; TP = power across all frequency bands (<0.4Hz); pVLF = power density in very low frequency band (<0.04 Hz); pLF = power density in low frequency band (0.04–0.15 Hz); pHF = power density in high frequency band (0.15–0.4 Hz); LF/HF= ratio between power in low frequency band to power in high frequency band; VLF= very low frequency band (<0.04 Hz); HF= high frequency band (0.15–0.4 Hz) ApEnHR = Approximate entropy for a heart rate time series; CEHR = Conditional entropy for a heart rate time series; CI= complexity index (The minimum of the CE with respect to L past value); SEHR = Sample entropy for a heart rate time series; RR is the corresponding to the time between consecutive R-waves in ECG and has a reciprocal relation with heart rate
Quality Assessment
The quality of included studies assessed using the Crombie criteria adapted for assessing cross-sectional studies [23]. Eight quality criteria, including research design, recruitment strategy, response rate, sample representativeness, objective and reliable measures, performed power calculation, proper statistical analysis and no evidence of bias were considered in the adapted criteria. Two more components of clarity and completeness in report and reproducibility of study were added to above-mentioned criteria, to assess the clarity of reported results and parameters and the reproducibility of the study. Higher overall score is an indicator of methodological quality in the study.
Data Analysis
Value of HR and HRV in frail participants was compared with non-frail and/or pre-frail participants, where possible. In the absence of these data, a summary of results was presented. Due to the lack of data and differences in study design and studied parameters, meta-analysis was not possible. Therefore, a narrative synthesis was employed to combine results and explain the conflicting results.
Results
Study Selection
We found 1903 articles through database searching and 12 additional articles through citation analysis of the selected articles and perusal of the World Wide Web. Of the 1544 articles remaining after duplicates were removed, 1490 were excluded during the title and abstract screening because of irrelevance to the topic (Figure 1). Inclusion/exclusion criteria as outlined above were applied to the full text of 54 articles and 48 articles were excluded due to the following reasons: 1) lack of a standard definition of frailty (n=31); 2) lack of HR or HRV measurements (n=8); 3) were conference abstract (n=5); 4) were non-English articles (n=3); and 5) was a systematic review (n=1). Finally, six articles [12, 13, 24–27] met all inclusion criteria (Figure 1).
Study Design and Quality Assessment
All included articles were cross-sectional studies, with various research aims with respect to cardiac ANS, as summarized in Table 1.
Using adapted Crombie criteria, study conducted by Romero-Ortuno et al. [25] had the highest score (8 out of 10) among all included studies. As reported in Table 2, all studies had an appropriate research design, appropriate recruitment strategy, objective and reliable measures, and were reproducible study. Only in Takahashi et al. study statistical analysis was not performed to assess pairwise differences between non-frail, pre-frail, and frail groups. Interestingly, none of the included studies performed power calculation. Katayama et al. was the only study that reported response rate [24]. Four research studies lack a sufficient report of parameters [12, 13, 24, 27]; statistical value (e.g., mean and standard deviation) for extracted parameters was not reported for these four articles [12, 13, 24, 27], and in three articles demographic information was not reported individually for each frailty group [12, 13, 27].
Table 2.
Quality assessment of articles using adapted Crombie criteria
| Author and year | Appropriate Research Design |
Appropriate Recruitment Strategy |
Response Rate, [%] |
Sample Representative |
Objective and Reliable Measures |
Power Calculation/ Justification of Numbers |
Appropriate Statistical Analysis |
No Evidence for Bias |
Clarity and completeness in report |
Reproducibility of study |
Quality Indicators Met |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Chaves et al. 2008 | ✓ | ✓ | NR | ✗* | ✓ | ✗ | ✓ | ✓ | ✗▲ | ✓ | 6/10 |
| Varadhan et al. 2009 | ✓ | ✓ | NR | ✗* | ✓ | ✗ | ✓ | ✓ | ✗▲ | ✓ | 6/10 |
| Weiss et al. 2009 | ✓ | ✓ | NR | ✗* | ✓ | ✗ | ✓ | ✓ | ✗▲▲ | ✓ | 6/10 |
| Romero-Ortuno et al. 2011 | ✓ | ✓ | NR | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | 8/10 |
| Takahashi et al. 2014 | ✓ | ✓ | NR | ✓ | ✓ | ✗ | ✗# | NC | ✓ | ✓ | 6/10 |
| Katayama et al. 2015 | ✓ | ✓ | 64% | ✗** | ✓ | ✗ | ✓ | ✓ | ✗▲▲▲ | ✓ | 7/10 |
NR: Not Reported; NC: Not Clear
Only elderly women with severe to moderate disability were included.
Only elderly women were recruited.
No suitable statistical test was used to assess pair-wise differences between non-frail, pre-frail and frail groups.
Mean and standard deviation of extracted parameters for each group of non-frail and frail was not reported. Also, demographic information across different groups (e.g. non-frail and frail) was not reported.
Mean and standard deviation of extracted parameters for each group of non-frail and frail were not reported. Also, number of participants and corresponding demographic information for two studied groups were not reported.
No report for the value of different studied heart rate derived parameters were reported across different frailty groups.
Participants
Sample sizes among selected articles ranged from 23 to 547, and participants’ ages ranged from 60 to 101 years. Total of 919 non-frail and 305 frail were studied among selected articles (excluding Weiss et al. study [27], within which the number of frail and non-frail participants was not provided). Interestingly, in four out of the six studies only female participants were recruited [12, 13, 24, 27]. Three of these studies [12, 13, 27] retrospectively analyzed different data subsets from Women’s Health and Aging Study (WHAS) which was originally designed for investigation of the epidemiology of disability progression in non-institutionalized women aged 65 and older [28]. Katayama et al. [24] did not explain the reason for recruiting only women.
In three studies [12, 13, 27] participants were categorized into non-frail and frail, and in two studies [24, 26] into non-frail, pre-frail, and frail according to Fried’s criteria. In one study [25] participants were categorized into three categories: non-frail, pre-frail and frail, using a modified version of the Fried’s criteria. Since some measures based on Fried’s criteria were not available in this study, they were substituted with similar measures extracted from the experiment (e.g., frequency of going outdoor was used as a surrogate of Minnesota Leisure Time Activity questionnaire used in Fried’s criteria) [25, 29]. Of note, adjustment of evaluated outcomes such as gait speed and hand grip strength use age and sex adjusted norms in all studies in accordance with frailty criteria (Fried’s criteria or modified Fried’s criteria).
Equipment and Protocols Used for Assessment of Cardiac Autonomic Control
Table 1 summarizes the equipment and protocols that were used for the assessment of cardiac ANS. In one study no information about the equipment utilized for assessment of heart response was reported [27]; all others used one of the methods described below:
HR extraction from Electrocardiogram (ECG): In two studies [12, 13] two-channels Holter monitoring, and in one study [30] 12-channels ECG recording equipment were utilized. In one study [24] a one channel ECG monitor (a belt worn around the thorax) was used; using this device beat-to-beat HR from ECG was recorded without providing the raw ECG data.
Optical HR Monitoring: One study [25] used an optical method based on measurement of changes in light absorption to estimate HR.
Weiss et al. [27] and Romero-Ortuno et al. [25] measured HR while participants were performing the seated step test, and the lying to standing orthostatic test, respectively. Takahashi et al. [30] assessed the participants during supine and standing postures (10 minutes in each position). Measurement of ECG in supine position for 10 minutes was done by Katayama et al. [24]. In the other two studies, participants were free to perform any physical tasks such as walking, lying, sitting, or standing [12, 13].
Between reviewed articles, HR was monitored during the clinical test by Weiss et al. [27] and during the test and three minutes after the test by Romero-ortuno et al. [25]. Takahashi et al. reported HRV parameters based on stationary sequences of 256 heart beats during each condition of supine and standing [30]. Katayama et al. used stationary five minutes period for reporting HR and HRV measures [24]. Furthermore, Chaves et al. and Varadhan et al. reported HRV based on 2–3 hours of recorded data [12, 13].
HR and HRV Parameters Reported
Definition of HR and HRV measures and their physiological interpretation are summarized in Appendix C. Detailed information about their calculation can be found in the provided reference.
As summarized in Table 3, two authors reported only HR parameters [25, 27], three studies reported only HRV measures [12, 13, 26], and one study provided both HR and HRV measures [24]. Of the three studies that measured HR [24, 25, 27], different outcome measures were calculated including baseline HR [25, 27], mean of beat-to-beat intervals (RR intervals) which has a reciprocal relation with HR [24] and changes in HR (delta HR) with different definitions as explained in Table 3 [25, 27].
Among studies that provided HRV assessment, outcome measures include: 1) time domain HRV measures: standard deviation of normal to normal intervals (SDNN) [12, 13, 24], square root of the mean squared differences between successive RR intervals (RMSSD) [12, 13, 24], standard deviation of the average normal to normal intervals (SDANN) [12], and percentage of successive RR intervals with differences larger than 50 millisecond (pNN50) [12], (see Appendix C for definitions); 2) frequency domain measures of HRV derived using Fast Fourier Transform: power density in very low frequency band [12, 13], low frequency band [12, 13, 24] and high frequency band [12, 13, 24] and the ratio of low frequency band to high frequency band (LF/HF) [12, 13, 24], and total power density [12]); and 3) non-linear measures of HRV: approximate entropy [12, 30], conditional entropy [30], sample entropy [24] as measures of cardiac ANS complexity as well as symbolic dynamics [24]. Definition of frequency bands and non-linear measures are presented in Appendix C.
Alterations in HR and HRV parameters with frailty
A summary of HR derived parameters values across frailty groups or a summary of findings are presented in Table 4. Chaves et al. demonstrated that the approximate entropy (the amount of regularity and the unpredictability of fluctuations over time) of HR is significantly less in frail compared to non-frail adults [12]. Also, a significant reduction of SDNN, SDANN, total power density, pVLF, pLF and LF/HF in individual with higher probability of frailty were reported by Chaves et al. [12]. Varadhan et al. proposed that an aggregate of HRV measures by second principal component analysis had an stronger association with frailty because it captures the impairment in cardiac ANS while discarding any redundancies in the highly correlated HRV parameters [13]. Approximate entropy marked differences between non-frail and pre-frail/frail groups in Takahashi et al. study [30]. They reported a higher approximate entropy in pre-frail and frail individuals compared to non-frails, which is conflicting with results of Chaves et al. study [12]. Katayama et al. reported a reduction trend in SDNN and RMSSD from non-frail to pre-frail and frail [24]. The assessment of HRV in frequency domain (pLF and pHF) and non-linear domain (symbolic analysis) highlighted a higher sympathetic and lower parasympathetic modulation in frail when compared to non-frail and pre-frail groups [24]. A non-significant trend towards a lower sample entropy in pre-frail and frail groups compared to non-frail group was observed in study performed by Katayama and colleagues [24]. Also, they have shown that mean HR during supine position is higher in frail compared to pre-frails and non-frails.
The analysis of HR in response to activity during a clinical test demonstrated a higher baseline HR in frail group compared to non-frail group [25, 27]. Also, in both these studies, the change in HR in response to activity (Delta HR) was less in the frail group [25, 27]. Romero-ortuno et al. also reported a significant decrease in maximum HR by 30 seconds after orthostatic lying to standing in frail compared to pre-frail and non-frail subjects [25].
Discussion
The purpose of the present systematic review was to study the relationship between the frailty syndrome and cardiac ANS. The results demonstrated a reduction in HRV and complexity and impairment in HR response to physical activity in the frail group compared to the non-frail group. These observations provide evidence of cardiac ANS impairments in the frail group compared to non-frails.
Measurement equipment
Findings from the current review revealed that HR extracted from ECGs recorded by ECG monitoring equipment (e.g., holter monitor, standard 12 lead ECG, and chest worn ECG monitoring equipment) was the most common approach for assessing HR in older adults. Optical HR monitoring was also employed for recording HR. Among these two approaches, using an ECG device that reports both HR and raw ECG data that used for extraction of HR is more efficient for excluding noisy data segments and heart beats originating from an ectopic location and/or arrhythmia and noisy data.
HR and HRV measures for assessing frailty
Of different HR measures, baseline HR and delta HR were the most commonly used parameters. Delta HR was the most effective parameter that significantly differentiate frail from non-frail groups. Furthermore, SDNN, RMSSD and power density at different frequency bands (e.g., very low frequency, low frequency, and high frequency) were the main parameters that were used to study HRV, which showed reduction in global HRV and imbalance in activity between sympathetic and parasympathetic nervous system in the frail group compared to the non-frail group. Different definitions of entropy (e.g., approximate entropy, conditional entropy, and sample entropy) were also reported for assessing complexity underlying cardiac ANS. These measures showed a reduction in complexity of cardiac ANS in frail compared to non-frail groups for long-term measurement.
Conflicting Results across Studies
Two conflicting results between included studies in the current systematic review were observed:
Increased pLF and LF/HF, and decreased pHF was observed in the frail group compared to the non-frail group in the study by Katayama et al. [24], which was not in line with findings of Chaves et al. [12].
Increased approximate entropy (an indicator of increase in complexity of cardiac ANS) in the frail group compared to non-frail and pre-frail groups reported by Takahashi et al. [26] disagreed with results provided by Chaves et al. [12].
The main source for above mentioned disagreement may be related to implemented methodological approaches; Takahashi et al. [26] and Katayama et al. [24] measures were based on short-term HR series (~ 5 minutes), while participants were in controlled static posture (e.g., supine posture). On the other hand, Chaves et al. [12] calculated the same parameters for longer data recordings (2–3 hour) during physical tasks with diverse postures (e.g., lying, sitting, and standing).
Clinical Implications
Results of the current review demonstrated that most HRV measures (e.g., SDNN, TP, pVLF, pLF, LF/HF, approximate entropy) declined with frailty. The reduction in SDNN and total spectral power in the frail group compared to the non-frail group is an indicator of reduced global HRV in the frail group. Furthermore, different LF/HF in frail compared to non-frail individuals is an indicator of an increased imbalance between the sympathetic and parasympathetic nervous systems [12, 13, 24]. The reduction of entropy measures (approximate entropy [12] and sample entropy [24]) reported in frail individuals suggests a reduction in physiological complexity underlying HR dynamics. This finding is in agreement with previous research that demonstrated a reduction in complexity of HR with aging [30, 31]. The reduction in complexity may be associated with an impaired interaction between subsystems and regulatory mechanism and was observed in pathological conditions [30]. Therefore, it can be concluded that the frail group is at a higher risk of cardiac disease, and better medical care is required for them to prevent cardiac complications.
The significant smaller alterations in HR in response to a postural transition in the frail group [25, 27] indicates that the cardiac system’s instantaneous response is less robust and competent in frails compared to non-frails. This finding suggest that, frail people especially those with heart problems are more susceptible to adverse health conditions when performing strenuous tasks, such as climbing the stairs; therefore, more rest between performing more active activities is required.
Previous studies have demonstrated that prehabilitation or rehabilitation prior to, or after, cardiac surgery leads to fewer post-operative complications and shorter post-operative length of stay [32, 33]. The clinical findings of the current systematic review provide evidence that using detectable alterations of the cardiac ANS can assist in clinical monitoring. Cardiac ANS assessment may be used, specifically in cardiac patients, as a complementary measurement in addition to current subjective/semi-subjective methods (e.g., Fried’s criteria) or walking-based frailty assessment methods (e.g., 5-m gait speed) that assess physical performance and are the most common frailty assessment approaches in cardiovascular care for older adults [7]. As such, monitoring of HRV measures may serve as a potential tool to guide prescriptive prehabilitation and rehabilitation programs. Further, by extension, this may offer additional benefits for reducing hospital stay and health care costs, and improving the quality of life in older adults.
Future Research Directions
The small number of studies included in this systematic review highlights the fact that published research on the association of frailty and cardiac autonomic control is limited. Also, the variety of employed protocols and parameters for assessing cardiac ANS, suggests a lack of standardized protocols for the assessment of cardiac autonomic control. Therefore, suggesting a guideline for standardized assessment of cardiac ANS in frail older adults would be helpful for cross-study comparisons.
Although different HRV measures, including time domain, frequency domain, and non-linear approaches have been used to study cardiac ANS across frailty groups, no study has quantitatively evaluated the relationship between the degree of autonomic dysfunction (the range of HR and HRV) with the frailty level. Providing such information in future studies could help to better understand the underlying mechanism of HR and HRV alterations with frailty. Since each HR and HRV measure assess cardiac ANS from specific stand point, extraction of different HR and HRV measures can better characterize cardiac ANS and increase the confidence in interpreting the results. For example both RMSSD and pHF carry information about activity of parasympathetic nervous system, and implementing both parameters may increase the confidence in interpreting the effect of frailty on parasympathetic nervous system.
Interestingly, only one study monitored HRV in male older adults [26] and another study that provided cardiac ANS assessment on both males and females only focused on monitoring of HR response and HR recovery [25]. Therefore, further study of the effect of frailty on cardiac ANS in males is warranted.
From four studies that utilized a standardized definition of frailty for assessing HRV, two of them studied short-term HRV measures (5 minute measure) [24, 26] and the other two assessed HRV measures for a longer duration (2–3 hour) [12, 13]. Future studies are warranted to evaluate the agreement between short-term and long-term HRV assessment in frail populations, as well as finding the most reliable approach for assessing HRV in frail older adults.
All reviewed articles, studied the heart response and cardiac autonomic control during static postures (e.g., sitting and standing) [24, 26] or without considering posture or the level of activity [12, 13]. Previous studies recommended that frailty can be identified with greater precision by studying the heart response to physical activity (e.g., sit to stand) [25, 27]. Wearable technologies can provide a tool to simultaneously measure physiological data (e.g., ECG) and kinematic data (e.g., body posture using accelerometers), which can then be used to assess HR response and recovery to tasks such as sit-to-stand or lie-to-stand during activities of daily living. Also, this coupling of activity and physiologic response could help to better study the heart response and recovery during different phases of studied activity (e.g., the leaning forward part of sit to stand, or transition from leaning position to standing).
No study to date has explored longitudinal changes in the performance of the cardiac autonomic control in response to frailty progression. Future longitudinal studies are recommended to explore this relationship. Also, predictive value of the cardiac ANS in combination to current frailty assessment methods such as Fried’s criteria for prediction of prospective frailty status change is another area that could be explored.
Previous studies in the area of aging has shown that HRV measures were more pronounced during the night [34]; these results warrant research studies to assess the cardiac ANS in individuals with different frailty status, defined by standardized definition during night time sleeping. It should be noted that the effect of confounders such as movement and change in posture can be minimized during sleep.
Further, the effect of exercise on the cardiac ANS in frail individuals has been not explored previously. Designing studies to systematically explore the effect of exercise on the cardiac ANS based on objective frailty levels is worth pursuing.
Limitations
With regard to cardiac autonomic functions such as HR and HRV in the frail group defined by a standard definition, there is a limited body of evidence on this topic. In addition to the scarcity of articles, two factors limited the comparability among included articles: 1) different parameters were used for assessing cardiac ANS; and 2) confounders such as posture, [35, 36] and level of activity, [37, 38] which may influence HR derived parameters and outcomes, were not controlled in all studies. Also, due to the above factors as well as missing values [12, 13, 24, 27], we were unable to perform a meta-analysis. Finally, different subsets from the same sample were used in three out of the six included study [12, 13, 27] reducing the total number of unique participants and their independence, thus the results need to be interpreted with caution.
Conclusion
The current systematic review provides evidence of cardiac autonomic nervous impairments in frail compared to non-frail older adults. This impairment is characterized by a reduction in complexity of heart rate dynamics, reduction in heart rate variability, or impaired heart rate response to daily physical activities. The lack of methodological rigour limits the generalizability of study findings (i.e., different measures and study designs) thus more rigorous studies, incorporating systematic measures and methods are required to better assess older adult cardiac autonomic nervous system associations on differing frailty status.
Acknowledgments
This study was partially supported by an STTR-Phase II Grant (Award No. 2R42AG032748) from the National Institute on Aging and Arizona Center on Aging’s Hudson Family. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Aging or the National Institutes of Health.
Appendix A
Table A-1.
PRISMA 2009 checklist adapted from Moher et al. [22]
| Section/topic | # | Checklist item | Reported on section |
|---|---|---|---|
| TITLE | |||
| Title | 1 | Identify the report as a systematic review, meta-analysis, or both. | Article Title |
| ABSTRACT | |||
| Structured summary | 2 | Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number. | Abstract |
| INTRODUCTION | |||
| Rationale | 3 | Describe the rationale for the review in the context of what is already known. | Introduction |
| Objectives | 4 | Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design (PICOS). | Introduction |
| METHODS | |||
| Protocol and registration | 5 | Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including registration number. | – |
| Eligibility criteria | 6 | Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale. | Methods-Inclusion/exclusion Criteria |
| Information sources | 7 | Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched. | Methods- Search Strategy |
| Search | 8 | Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated. | Appendix B |
| Study selection | 9 | State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis). | Methods- Study Selection |
| Data collection process | 10 | Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators. | – |
| Data items | 11 | List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made. | Methods- Data Extraction |
| Risk of bias in individual studies | 12 | Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis. | – |
| Summary measures | 13 | State the principal summary measures (e.g., risk ratio, difference in means). | N/A |
| Synthesis of results | 14 | Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I2) for each meta-analysis. | N/A |
| Risk of bias across studies | 15 | Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies). | N/A |
| Additional analyses | 16 | Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified. | N/A |
| RESULTS | |||
| Study selection | 17 | Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram. | Results- Study Selection; Figure 1 |
| Study characteristics | 18 | For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations. | Results- Participant and Results- Study Design and Quality Assessment, Tables 1, 3, and 4 |
| Risk of bias within studies | 19 | Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12). | N/A |
| Results of individual studies | 20 | For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot. | N/A |
| Synthesis of results | 21 | Present results of each meta-analysis done, including confidence intervals and measures of consistency. | N/A |
| Risk of bias across studies | 22 | Present results of any assessment of risk of bias across studies (see Item 15). | N/A |
| Additional analysis | 23 | Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]). | N/A |
| DISCUSSION | |||
| Summary of evidence | 24 | Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers). | Discussion |
| Limitations | 25 | Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias). | Discussion- Limitations |
| Conclusions | 26 | Provide a general interpretation of the results in the context of other evidence, and implications for future research. | Conclusion |
| FUNDING | |||
| Funding | 27 | Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review. | Acknowledgements |
Appendix B
The following search strategy was used in the PubMed database:
“Frail Elderly”[Mesh] OR frail*[tw] OR pre-frail[tw] OR pre-frailty[tw] OR prefrail*[tw]
AND
((heart OR cardiac[Text Word])) AND (rate OR rhythm OR conduction[Text Word]) OR ((adrenerg* OR noradrenerg* OR cholinerg*)) OR ((autonomic[Text Word]) AND (control OR system[Text Word])) OR ((sympathetic OR parasympathetic[Text Word])) OR ((vasovagal OR vagal OR vagus[Text Word])) OR (((“Hemodynamics”[Mesh]) OR “Heart Conduction System”[Mesh]) OR “Autonomic Nervous System”[Mesh])
Search strategies applied in the other databases (Embase, Cochrane Library, Web of Science, CINAHL and ClinicalTrials.gov) were derived from the PubMed search. The database searches were conducted without using language or publication date restrictions (though non-English articles were excluded during the screening process).
Appendix C
Table C-1.
Definition of heart rate and heart rate variability (HRV) parameters and the physiological interpretation of HRV measures
| Parameter | Definition | Physiological Interpretation | |
|---|---|---|---|
| Heart Rate | Baseline HR, [BPM] | The HR before starting the test [16, 18] | — |
| Delta HR, [BPM] or [%] | Change in HR compared to the baseline after completing a test [16, 18] | — | |
| Heart Rate Variability | SDNN, [ms] | Standard deviation of RR intervals [35] | global HRV |
| SDANN, [ms] | Standard deviation of the average normal to normal (N-N) intervals over 5-minute periods [35] | sympathetic and parasympathetic nervous system interactions (autonomic tone) | |
| RMSSD, [ms] | Square root of the mean squared differences between successive RR intervals [35] | parasympathetic activity | |
| pNN50, [%] | Number of successive RR intervals that with differences larger than 50 ms divided by total number of RR intervals [35] | parasympathetic activity | |
| Total power density, [ms2] | Power across all frequency bands (<0.4Hz) which is a net effect of all possible physiological mechanism contributing in HRV [10, 36] | global HRV | |
| pVLF, [ms2 ] | Power in very low frequency band (<0.04Hz) [10, 36] | overall activity of various slow mechanisms of sympathetic function | |
| pLF, [ms2 ] | Power in low frequency band (0.04–0.15Hz) [10, 36] | both sympathetic and parasympathetic tone | |
| pHF, [ms2 ] | Power in high frequency band (0.15–0.4Hz) [10, 36] | parasympathetic tone | |
| LF/HF | Ratio between power in low frequency band to power in high frequency band[10, 36] | balance between sympathetic and parasympathetic tone | |
| Approximate Entropy (ApEn) Conditional Entropy (CE) Sample Entropy (SE) |
Measures the irregularity or complexity of the HR series [34] | — | |
| Symbolic Analysis | Convert cardiac RR intervals into different levels (symbols) and analyse pattern of symbols | — |
RR Intervals: the time interval between adjacent QRS complex in ECG; normal to normal (N-N) intervals: used to emphasize RR intervals that are normal (not ectopic beats); HRV: heart rate variability
Footnotes
Conflict of Interest
The authors have no conflicts of interest to disclose.
References
- 1.Ferrucci L, et al. Biomarkers of frailty in older persons. Journal of endocrinological investigation. 2001;25(10 Suppl):10–15. [PubMed] [Google Scholar]
- 2.Fried LP, et al. Frailty in older adults evidence for a phenotype. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2001;56(3):M146–M157. doi: 10.1093/gerona/56.3.m146. [DOI] [PubMed] [Google Scholar]
- 3.Boyd CM, et al. Clinical practice guidelines and quality of care for older patients with multiple comorbid diseases: implications for pay for performance. Jama. 2005;294(6):716–724. doi: 10.1001/jama.294.6.716. [DOI] [PubMed] [Google Scholar]
- 4.FugateWoods N, et al. Frailty: emergence and consequences in women aged 65 and older in the Women’s Health Initiative Observational Study. Journal of the American Geriatrics Society. 2005;53(8):1321–1330. doi: 10.1111/j.1532-5415.2005.53405.x. [DOI] [PubMed] [Google Scholar]
- 5.Afilalo J, et al. Frailty Assessment in the Cardiovascular Care of Older Adults. Journal of the American College of Cardiology. 2013 doi: 10.1016/j.jacc.2013.09.070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Flint KM, et al. Frailty and the Selection of Patients for Destination Therapy Left Ventricular Assist Device. Circulation-Heart Failure. 2012;5(2):286–U311. doi: 10.1161/CIRCHEARTFAILURE.111.963215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Afilalo J, et al. Frailty assessment in the cardiovascular care of older adults. Journal of the American College of Cardiology. 2014;63(8):747–762. doi: 10.1016/j.jacc.2013.09.070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Afilalo J, et al. Role of frailty in patients with cardiovascular disease. The American journal of cardiology. 2009;103(11):1616–1621. doi: 10.1016/j.amjcard.2009.01.375. [DOI] [PubMed] [Google Scholar]
- 9.Sergi G, et al. Pre-frailty and risk of cardiovascular disease in elderly men and women: the Pro. VA study. Journal of the American College of Cardiology. 2015;65(10):976–983. doi: 10.1016/j.jacc.2014.12.040. [DOI] [PubMed] [Google Scholar]
- 10.Jänig W. Integrative action of the autonomic nervous system: Neurobiology of homeostasis. Cambridge University Press; 2006. [Google Scholar]
- 11.Freitas J, et al. Role of autonomic nervous system and hemodynamics in cardiovascular homeostasis after orthostatic stress. Revista portuguesa de cardiologia: orgao oficial da Sociedade Portuguesa de Cardiologia= Portuguese journal of cardiology: an official journal of the Portuguese Society of Cardiology. 2000;19(12):1241–1274. [PubMed] [Google Scholar]
- 12.Chaves PH, et al. Physiological Complexity Underlying Heart Rate Dynamics and Frailty Status in Community-Dwelling Older Women. Journal of the American Geriatrics Society. 2008;56(9):1698–1703. doi: 10.1111/j.1532-5415.2008.01858.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Varadhan R, et al. Frailty and impaired cardiac autonomic control: new insights from principal components aggregation of traditional heart rate variability indices. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2009;64(6):682–687. doi: 10.1093/gerona/glp013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Uchmanowicz I, et al. Frailty in Heart Failure. Current heart failure reports. 2014:1–8. doi: 10.1007/s11897-014-0198-4. [DOI] [PubMed] [Google Scholar]
- 15.Samala RV, et al. Heart failure in frail, older patients: We can do ‘MORE’. Cleveland Clinic Journal of Medicine. 2011;78(12):837–845. doi: 10.3949/ccjm.78a.11085. [DOI] [PubMed] [Google Scholar]
- 16.Olshansky B, et al. Parasympathetic nervous system and heart failure pathophysiology and potential implications for therapy. Circulation. 2008;118(8):863–871. doi: 10.1161/CIRCULATIONAHA.107.760405. [DOI] [PubMed] [Google Scholar]
- 17.Triposkiadis F, et al. The sympathetic nervous system in heart failure: physiology, pathophysiology, and clinical implications. Journal of the American College of Cardiology. 2009;54(19):1747–1762. doi: 10.1016/j.jacc.2009.05.015. [DOI] [PubMed] [Google Scholar]
- 18.Acharya UR, et al. Heart rate variability: a review. Medical and Biological Engineering and Computing. 2006;44(12):1031–1051. doi: 10.1007/s11517-006-0119-0. [DOI] [PubMed] [Google Scholar]
- 19.Nicolini P, et al. The prognostic value of heart rate variability in the elderly, changing the perspective: from sympathovagal balance to chaos theory. Pacing and clinical electrophysiology. 2012;35(5):621–637. doi: 10.1111/j.1540-8159.2012.03335.x. [DOI] [PubMed] [Google Scholar]
- 20.Sajadieh A, et al. Increased heart rate and reduced heart-rate variability are associated with subclinical inflammation in middle-aged and elderly subjects with no apparent heart disease. European Heart Journal. 2004;25(5):363–370. doi: 10.1016/j.ehj.2003.12.003. [DOI] [PubMed] [Google Scholar]
- 21.Tsuji H, et al. Impact of reduced heart rate variability on risk for cardiac events The Framingham Heart Study. Circulation. 1996;94(11):2850–2855. doi: 10.1161/01.cir.94.11.2850. [DOI] [PubMed] [Google Scholar]
- 22.Moher D, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of internal medicine. 2009;151(4):264–269. doi: 10.7326/0003-4819-151-4-200908180-00135. [DOI] [PubMed] [Google Scholar]
- 23.Petticrew M, Roberts H. Systematic reviews in the social sciences: A practical guide. John Wiley & Sons; 2008. [Google Scholar]
- 24.Katayama PL, et al. Cardiac autonomic modulation in non-frail, pre-frail and frail elderly women: a pilot study. Aging clinical and experimental research. 2015:1–9. doi: 10.1007/s40520-015-0320-9. [DOI] [PubMed] [Google Scholar]
- 25.Romero-Ortuno R, et al. Orthostatic haemodynamics may be impaired in frailty. Age and ageing. 2011;40(5):576–583. doi: 10.1093/ageing/afr076. [DOI] [PubMed] [Google Scholar]
- 26.Takahashi A, et al. Cardiovascular Oscillations (ESGCO), 2014 8th Conference of the European Study Group on. IEEE; 2014. Short-term complexity of cardiovascular oscillations in frailty syndrome. [Google Scholar]
- 27.Weiss CO, et al. Relationships of cardiac, pulmonary, and muscle reserves and frailty to exercise capacity in older women. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2010;65(3):287–294. doi: 10.1093/gerona/glp147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Guralnik JM, et al. The Women’s Health and Aging Study: health and social characteristics of older women with disability. DIANE Publishing; 1995. [Google Scholar]
- 29.Romero Ortuno R. Orthostatic hypotension as a marker of frailty in older people. Trinity College Dublin; 2011. [Google Scholar]
- 30.Takahashi AC, et al. Aging reduces complexity of heart rate variability assessed by conditional entropy and symbolic analysis. Internal and emergency medicine. 2012;7(3):229–235. doi: 10.1007/s11739-011-0512-z. [DOI] [PubMed] [Google Scholar]
- 31.Beckers F, Verheyden B, Aubert AE. Aging and nonlinear heart rate control in a healthy population. American Journal of Physiology-Heart and Circulatory Physiology. 2006;290(6):H2560–H2570. doi: 10.1152/ajpheart.00903.2005. [DOI] [PubMed] [Google Scholar]
- 32.Carli F, Zavorsky GS. Optimizing functional exercise capacity in the elderly surgical population. Current Opinion in Clinical Nutrition & Metabolic Care. 2005;8(1):23–32. doi: 10.1097/00075197-200501000-00005. [DOI] [PubMed] [Google Scholar]
- 33.Stewart KJ, et al. Cardiac rehabilitation following percutaneous revascularization, heart transplant, heart valve surgery, and for chronic heart failure. Chest Journal. 2003;123(6):2104–2111. doi: 10.1378/chest.123.6.2104. [DOI] [PubMed] [Google Scholar]
- 34.Mourey F, et al. Assessment of cardiac autonomic nervous activity in frail elderly people with postural abnormalities and in control subjects. Archives of gerontology and geriatrics. 2009;48(1):121–124. doi: 10.1016/j.archger.2007.11.004. [DOI] [PubMed] [Google Scholar]
- 35.Martinmäki K, et al. Intraindividual validation of heart rate variability indexes to measure vagal effects on hearts. American Journal of Physiology-Heart and Circulatory Physiology. 2006;290(2):H640–H647. doi: 10.1152/ajpheart.00054.2005. [DOI] [PubMed] [Google Scholar]
- 36.Walawalkar SR. A Study Of Variation In Heart Rate Variability With Change In Posture In Young Adult Indian Males [Google Scholar]
- 37.Perini R, Veicsteinas A. Heart rate variability and autonomic activity at rest and during exercise in various physiological conditions. European journal of applied physiology. 2003;90(3–4):317–325. doi: 10.1007/s00421-003-0953-9. [DOI] [PubMed] [Google Scholar]
- 38.Sun FT, et al. Mobile Computing, Applications, and Services. Springer; 2012. Activity-aware mental stress detection using physiological sensors; pp. 211–230. [Google Scholar]