Emerging Technologies and Methods of Assessing Heart Rate Variability to Determine Cardiovascular Disease Risk

Link to google drive doc with sources.
Warning: This is a research paper and will take an estimated 10-15 minutes to read.

In recent decades heart rate variability (HRV) has been applied in the clinical setting to treat and diagnose cardiovascular disease and other disorders. Extensive research into HRV can be traced to the early 1990s. HRV involves measuring the beat-to-beat variability by the heart to assess the function of the autonomic nervous system. HRV can therefore be used as a tool to assess the function of the parasympathetic nervous system or vagal tone as well. Many different mathematical equations or logarithms currently exist to measure and express HRV. The literature most commonly expresses HRV as the standard deviation of N-N intervals (SDNN) or as the square root of the mean of the sum of the squares of the successive differences between adjacent normal R-R intervals (RMSSD). Observing the variability from a low frequency gives insight into sympathetic modulation while high frequency measures reflect vagal modulation (Pinhel, et. al, 2016). A low variability is associated with increased risk of cardiovascular disease as well as an increase in all cause mortality. Inversely, a high variability is associated with better autonomic function and health status. A high variability is therefore desirable as it demonstrates ability of the autonomic nervous system to properly regulate the cardiovascular system. Common methods of measuring HRV include: 24-hour holter monitoring, ECG, plethysmography or photoplethysmography. Recently, there have been novel methods introduced to measure and track HRV scores. Smart-phone technology has been developed to measure and track HRV remotely without the need to enter a doctor’s office. Methods for smart-phone HRV testing range from utilization of photoplethysmography and certain ECG adapters that connect directly to smartphone applications. Certain applications function to detect arrhythmias as well and can immediately direct any information gathered by an application on the smart phone to the individual’s physician or health care providers.

            To begin with, cardiovascular disease contributes a large sum to overall health care spending in the United States and most industrialized nations. In terms of costs, cardiovascular disease accounted for $316.1 billion in the United States between 2012-2013 and included $189.7 billion for direct medical expenses and $126.7 billion in lost productivity costs (“Chronic Disease Prevention and Health Promotion”, 2017). The total cost of cardiovascular disease is double of the cost associated with treating cancer which currently stands at over $150 billion in expenses. In addition to economic burden, cardiovascular disease is responsible for a large portion of cause of death in the United States. In fact, cardiovascular disease in most industrialized countries is the leading cause of death and morbidity for aging populations (Greiser, et. Al, 2009). The American Heart Association uses their “Life’s simple 7” to measure progress in reducing the prevalence of cardiovascular disease (“Heart and Stroke Statistics”). These 7 factors include: smoking, physical inactivity, nutrition, overweight and obesity, cholesterol, diabetes, and high blood pressure. These risk factors have been identified by the American Heart Association to be modifiable and therefore could provide great cardiovascular benefit when improved.

            Although HRV was not a factor included in the American Heart Association’s list, it can potentially be an effective tool in assessing, improving, and monitoring progression of certain cardiovascular diseases. Identification of cardiovascular abnormalities or disease is especially important for treatment. Many individuals with cardiovascular disease such as atherosclerosis are asymptomatic. This means that patients could not receive treatment until they begin to experience complications or a cardiovascular event. Therefore, identifying risk for future cardiac events and implementing lifestyle changes could not only save individuals on healthcare costs, but could improve quality of life and lifespan as well. Some of the risk factors listed in the Life’s Simple 7 can lead to a decrease HRV if not improved. Although, the study by Greiser et. al. indicates that there is a weak and inconsistent association between physical activity, smoking, or alcohol with HRV within the literature. The author does admit that many of the measurements are difficult to compare and are conflicting in many circumstances. This could be attributed to HRV being assessed via a large variety of methods, frequencies, and algorithms. Although, an inverse association with HRV was present between triglycerides, glucose, waist-to-hip ratio, and diabetes in both genders, and cholesterol and hypertension had only a consistent inverse association in men (Gresier, et. Al., 2009). Yet, Lampert et. al found there to be an association between inflammatory markers and decreased HRV. The author mentioned alterations in sympathetic and parasympathetic function were found to affect proinflammatory cytokines which would be reflected in HRV changes. Although inflammatory markers are not necessarily an indicator of heart disease, some risk factors traditionally associated with coronary artery disease can also be associated with altered autonomic function (Lampert, et. Al. 2008). Lampert and colleagues found that both CRP and IL-6 both correlated with very low frequency and ultra-low frequency HRV (Lampert, et. Al. 2008). This correlation was not present in a high frequency setting. This indicates a relationship between sympathetic modulation, HRV, and inflammatory markers. Some of the Life’s simple 7 such as smoking, obesity, physical inactivity, and hypertension were also found in this study to have an inverse relationship with HRV. These results conflict with the prior study where no association was found between smoking or physical inactivity. But, it is consistent regarding an association between obesity, hypertension in males, and HRV. Another assessment of autonomic function, left ventricular diastolic function, was studied by Habek et. Al. Left ventricular diastolic dysfunction was observed in 79% of type 2 diabetic patients studied and associated with decreased HRV (Habek, et. Al., 2014). Although diabetes is not a cardiovascular disease, it does share overweight or obesity as a common risk factor and a relationship with HRV. Diabetes is inversely associated with HRV as mentioned in the prior studies as well. 83% of those with left ventricular diastolic function were found to have a decreased HRV while only 7% of those with normal left ventricular diastolic function were found to have significantly decreased HRV (Habek, et. Al., 2014). A study by Ogliari Et. Al. indicated there to be a consistent association between high resting heart rate and low HRV in previous studies (Ogliari, et. Al, 2015). More cardiac autonomic control is necessary to provide adequate perfusion to the brain and other essential organs. The study indicated that the individuals with decreased HRV had higher levels of blood pressure variability. An increase in the variability of one’s blood pressure could lead to a higher mean or normalized blood pressure reading. This means that higher variability in blood pressure could be a potential sign or effect of hypertension. This thereby also exposes individuals to risks from hypertension such as atherosclerosis. A presence of atherosclerosis is detrimental to the flow of blood through the arteries. This detriment to blood flow increases risks for certain cardiovascular events such as a myocardial infarction or stroke. This study specifically examined the relationship between resting heart rate, HRV, and functional status of older adults. Functional stratus was assessed by activities of daily living (ADL). The researchers found that functional decline could be associated with decreased HRV independent of any comorbities, sex, or cardiovascular risks factors (Ogliari, et. Al., 2015). Reduced HRV and resting heat rate reflect altered balance between the parasympathetic and sympathetic nervous system. Having an over active sympathetic system has been linked to atherosclerotic risk factors such as obesity and subclinical inflammation (Ogliari, et. Al. 2015). Both obesity and inflammation were previously identified to have an association with HRV. The author closes by remarking that there should be more investigation into the effects of physical activity on functionality through improved autonomic regulation. The author also recommends investigation into if HRV could be a marker for indicating functional decline in older adults. In a study by Sloan Et. Al, the effects of aerobic training on autonomic regulation in young adults was studied. An initial limitation of the study is the studied population. There is a difference between young adults and the populations, middle aged and older adults, studied in the previous studies.  Although, the mechanism in which exercise improves autonomic function in any population can be compared. Providing early solutions and modifying risk factors such as physical inactivity early could prove to be a beneficial strategy to reducing the impact of cardiovascular disease. Researchers in this study found that in response to aerobic exercise aerobic capacity increased, VO2 max increased, resting heart rate decreased, and high frequency R-R interval variability (RRV) increased (Sloan, et. Al., 2009). High frequency measures for HRV more accurately reflect parasympathetic regulation therefore these results show autonomic regulation, primarily through vagal tone, improved from aerobic exercise. These results were not found in groups participating in strength training. These results were also more present in men than in women. The results from this study draw a few parallels with previous studies. Ogliari et. Al found an association between low resting heart rate and decreased HRV as well. Men in Greiser et. Al. showed a greater association as did the men in the Sloan et. Al. study. The author concludes that the evidence is consistent to prove that there is a cardioprotective effect from aerobic exercise and suggests that atherosclerosis has an origin in childhood (Sloan, et. Al., 2009). This is significant regarding early detection and prevention of cardiovascular disease. In another study conducted by Heitmann Et. Al. multivariate short-term HRV analysis was assessed as a screening tool for identifying the early stages of heart disease (Heitmann, et. Al. 2011). The researchers concluded that the method is potentially a useful pre-diagnostic tool for cardiovascular disease that can be utilized in the primary care setting (Heitmann, et. Al. 2011). This study utilized short term symbolic dynamics (STSD) which was a different method than in some of the previously discussed studies. This method differs in duration as well. A holter monitor is often utilized which analyzes the HRV of the patient for generally a 24-hour period. The method utilized in this study was 5 minutes in duration. This short-term reading could prove to be more practical and cost-effective for both practitioners and patients. Although the method in this study only determined whether a heart disease was currently present. The method was not able to identify which cardiovascular disease was present. Although this may seem like a major limitation, early identification of autonomic dysregulation could prove to be beneficial in treatment. The method could potentially reduce the number of false referrals by practitioners as well (Heitmann, et. Al. 2011). Consistent with the previous studies, the research did observe differences when gender and age were accounted for. The final article published by the annual review of medicine examined the relationship between HRV and various cardiovascular diseases including: myocardial infarctions, ventricular arrhythmias, and congestive heart failure. The author indicates that there is a consistent association between decreased HRV with coronary heart disease mortality, sudden death, and all-cause mortality (Stein, et. Al., 1999). In patients with myocardial infarction a lower HRV was connected to increased risk of mortality even after adjusting for risk factors such as ejection fraction (Stein, et. Al., 1999). The author mentions that although HRV was a better predictor than most risk factors, it is more clinically significant if combined with other risk factors. In terms of arrhythmias, the threshold for ventricular fibrillation was found to be lower with decreased parasympathetic and increased sympathetic tone. Patients with decreased heart rate variability were also then found to be at greater risk for sudden cardiac death (Stein, et. Al., 1999).  Malignant arrhythmias were found to have a similar association to heart rate variability scores as sudden death. The same change in parasympathetic and sympathetic tone could be observed in those with congestive heart failure. Lower HRV is also consistently present in those with congestive heart failure (Stein, et. Al., 1999). Although lower HRV was not consistently associated with increased mortality in those that have congestive heart failure. The author states that there is no evidence to suggest that increasing HRV would improve survival rates. Although, many interventions used to decrease mortality are associated with increased HRV (Stein, et. Al., 1999). This does not necessarily mean that quality of life cannot improve with improvement of HRV as well. In summary, some of the research methods in terms of frequency, time domains, and equipment differ. There is some evidence suggesting no relationship between HRV and its association with cardiovascular disease or increased mortality from cardiovascular events. The consensus from the bulk of the research and physiological understanding of HRV seem to indicate that a relationship is present between cardiovascular disease and decreased HRV.

            Traditionally, the electrocardiogram (ECG) and 24-hour holter monitoring have been used as assessment tools. Research methods vary in the tools or equipment they use to measure HRV. Walsh et. Al gives insight in an article about various new technologies being utilized to measure HRV. Recently, over 90% of Americans own a cell phone and about 55% own a smart phone (Walsh, et. Al. 2014). The author mentions how now medical practitioners are taking advantage of the expansive technology and digital network by integrating medical data and records with mobile devices. Different adapters and patches can save costs for customers as they link directly to their smart phones and can instantly provide data to their medical providers. This includes monitoring of vital signs, arrhythmia detection, and HRV analysis. A major question about these applications regards their accuracy of measurement. The research in this area is currently very limited as the availability of this technology was not common until recent years. A commonly applied technology for smart phones has been photoplethysmography (PPG). The technology was first developed in the 1960s and relies on the idea that tissue and blood have different properties of absorbing light (Russoniello, et. Al. 2010). The individuals finger is placed on a photocell and light is shined through. The amount of light reaching the cell is the measured and assessed to measure pulse volume or beat variations. The beat variability of peripheral blood flow is similar to beat variability as seen observed on an ECG (Russoniello, et. Al. 2010).  A study titled, “Can photoplethysmography variability serve as an alternative approach to obtain heart rate variability information?” examined the accuracy of this method. The authors were able to conclude that measures from PPG correlated very well (r=0.99 for supine SDNN and RMSSD) with ECG readings (Lu, et. Al, 2008). The device is much simpler to administer than an ECG and is less invasive. An advantage of this technology is that it provides more measures such as blood oxygenation level, HRV information, and respiratory rate (Lu, et. Al, 2008). The information could also be provided wirelessly. This would mean less office visits to practitioners would be necessary and an instant stream of information would be available to both the patient and practitioner. A similar study be Russoniello et. Al. found similar correlation (r=0.99 for SDNN and RMSSD) when comparing the ECG with PPG (Russoniello, et. Al., 2010). Although obese children were the population in this study, seeing validity for a tool across populations has importance. The author mentions as well that the method is cost effective and a potentially useful diagnostic tool for those with autonomic nervous system related conditions (Russoniello, et. Al., 2010). The center for disease control and prevention (CDC) currently estimates that about 1 in every 6 children in the United States are obese (“Heart and Stroke Statistics”). With the persistent high rates of childhood obesity, it is important to implement identification and treatment strategies early to minimize diminished cardiovascular health at older age. A study by Parra et. Al. investigated various smart phone applications and monitoring methods to assist with improving the quality of life for older adults. This article examined older adults in Europe as opposed to the United States. It was found that 85% of older adults had at least one chronic disease while 65% had two or more chronic disease (Parra, et. Al., 2016). Some of the most common conditions include hypertension and heart disease. 61% of men and 71% of women were found to be living alone (Parra, et. Al., 2016). This could pose a potential issue concerning sudden cardiac events such as a myocardial infarction or even a stroke. The article mentions utilizing a smart phone camera similarly to PPG to assess HRV. A high correlation (r= 0.9844 RMSSD) was found utilizing an iPhone4s to obtain accurate HRV (Lu, et. Al., 2008). Although this study was conducted on very few subjects (n=25) and therefore may need more investigation. The author concludes by remarking the potential reach of this kind of technology in the future will grow as the elderly population in many industrialized countries is growing. Limitations of this technology includes the individual to have the requisite functional capacity to use smart phone applications. Issues associated with old age such as loss of dexterity or loss of sight could prove to be a potential issue for some individuals. It was found in 2013 that only about 18% of Americans over the age of 65 currently use smart phones although this number is up significantly from 13% in February of 2012 (Parra, et. Al., 2016). This also possess a potential limitation as much of the population does not own or operate a smart phone. This is subject to change and expand as the population ages. Many younger individuals (ages 18-34) own and use smart phones with a user rate of over 80% (Parra, et. Al., 2016).

            In summary, HRV is used in the clinical field to assess autonomic function. There are various means of measuring HRV. The most common values measuring HRV include SDNN and RMSSD. The various algorithms and mathematical means of measuring HRV could prove to be a point of inconsistency and difficulty when attempting to compare research. Although some of the research indicates this inconsistency and manifests it within differing results, the preponderance of evidence seems to indicate that an association between HRV and cardiovascular disease exists. A low HRV in prior research indicated autonomic dysfunction and in some of the research correlated with presence of a cardiovascular disease. The inconsistency in some of the research can be explained by several variables. As mentioned earlier, standardized units of measurement are an issue, equipment use, and application. Devices such as the holter monitor, PPG, ECG, or a smart phone are all different technologies. Although, many have been validated in some manner with at least one other technology. More studies need to be conducted to examine the validity of each method in comparison to emerging technologies. Application refers to the use of HRV as a measure itself. Studies use HRV to correlate or differentiate amongst different variables or diseases. Some examined in this study included not just cardiovascular disease, but diabetes and inflammation. Since the amount of research in this field is rather expansive, it should come to no surprise that it has surpassed being isolated to just one area such as cardiovascular disease. This is also an important consideration to bare in mind when comparing the literature. Finally, the daily use and availability of technology allows us an unprecedented level of convenience and communication. By improving the technology many of us already hold in the palms of our hands we could reduce health care costs, provide more convenient health care, and provide more health monitoring. Providing patients with a convenient source of health information in their smart phones about their cardiovascular health in that very moment is invaluable. Currently cardiovascular disease sits atop the charts of cause of death for the United States and many industrialized countries. Early detection of autonomic dysfunction using self-administered technology, such as smart phone health applications, could potentially be a useful treatment and diagnostic tool for both the patient and practitioner. 

References

Greiser, K. H., Kluttig, A., Schumann, B., Swenne, C. A., Kors, J. A., Kuss, O., . . . Werdan, K. (2009). Cardiovascular diseases, risk factors and short-term heart rate variability in an elderly general population: The CARLA study 2002-2006. European Journal of Epidemiology, 24(3), 123-42. doi:http://library.semo.edu:2275/10.1007/s10654-009-9317-z

Habek, J. C., Lakusic, N., Kruzliak, P., Sikic, J., Mahovic, D., & Vrbanic, L. (2014). Left ventricular diastolic function in diabetes mellitus type 2 patients: Correlation with heart rate and its variability. Acta Diabetologica, 51(6), 999-1005. doi:http://library.semo.edu:2275/10.1007/s00592-014-0658-z

Heitmann, A., Huebner, T., Schroeder, R., Perz, S., & Voss, A. (2011). Multivariate short-term heart rate variability: A pre-diagnostic tool for screening heart disease. Medical and Biological Engineering and Computing, 49(1), 41-50. doi:http://library.semo.edu:2275/10.1007/s11517-010-0719-6

Kim, Y. J., Heo, J., Park, K. S., & Kim, S. (2016). Proposition of novel classification approach and features for improved real-time arrhythmia monitoring. Computers in Biology and Medicine, 75, 190-202. doi:http://library.semo.edu:2275/10.1016/j.compbiomed.2016.06.009

Lampert, R., Bremner, J. D., Su, S., Miller, A., Lee, F., Cheema, F., . . . Vaccarino, V. (2008). Decreased heart rate variability is associated with higher levels of inflammation in middle-aged men. The American Heart Journal, 156(4), 759.e1-7. doi:http://library.semo.edu:2275/10.1016/j.ahj.2008.07.009

Lehrer, P. M., & Gevirtz, R. (2014). Heart rate variability biofeedback: How and why does it work? Frontiers in Psychology, 5. doi:10.3389/fpsyg.2014.00756

Lu, S., PhD., Zhao, H., M.S., Ju, K., PhD., Shin, K., PhD., Lee, M., PhD., Shelley, K., PhD., & Chon, K. H., PhD. (2008). Can photoplethysmography variability serve as an alternative approach to obtain heart rate variability information? Journal of Clinical Monitoring and Computing, 22(1), 23-9. doi:http://library.semo.edu:2275/10.1007/s10877-007-9103-y

Ogliari, G., M.D., Mahinrad, S.,M.D.M.Sc, Stott, D. J.,M.D.PhD., Jukema, J. W., Mooijaart, S. P.,M.D.PhD., Macfarlane, P. W.,M.D.PhD., . . . Sabayan, B.,M.D.PhD. (2015). Resting heart rate, heart rate variability and functional decline in old age. Canadian Medical Association.Journal, 187(15), 1. Retrieved from https://library.semo.edu:2443/login?url=https://library.semo.edu:4836/docview/1725584846?accountid=38003

Parra, L., Sendra, S., Jiménez, J. M., & Lloret, J. (2016). Multimedia sensors embedded in smartphones for ambient assisted living and e-health. Multimedia Tools and Applications, 75(21), 13271-13297. doi:http://library.semo.edu:2275/10.1007/s11042-015-2745-8

Pinhel MAS, Oliveira BAP, Nicoletti CF, Noronha NY, Quinhoneiro DCG, Cortes-Oliveira C, Gardim CB, Júnior WS, Marchini JS, Souza DRS, Fernandes de Godoy M, Nonino CB (2016). Heart Rate Variability as a Tool for Predicting Risk of Cardiovascular Disease in Patients with Obesity. Acad. J. Sci. Res. 4(5): 117-123.

Russoniello, C. V., Pougtachev, V., Zhirnov, E., & Mahar, M. T. (2010). A measurement of electrocardiography and photoplethesmography in obese children. Applied Psychophysiology and Biofeedback, 35(3), 257-9. doi:http://library.semo.edu:2275/10.1007/s10484-010-9136-8

Sloan, R. P., PhD., Shapiro, P. A., M.D., DeMeersman, R. E., PhD., Bagiella, E., PhD., Brondolo, E. N., PhD., McKinley, P. S., PhD., . . . Myers, M. M., PhD. (2009). The effect of aerobic training and cardiac autonomic regulation in young adults. American Journal of Public Health, 99(5), 921-8. Retrieved from https://library.semo.edu:2443/login?url=https://library.semo.edu:4836/docview/215086726?accountid=38003

Stein, P. K., & Kleiger, R. E. (1999). Insights from the study of heart rate variability. Annual Review of Medicine, 50, 249-61. Retrieved from https://library.semo.edu:2443/login?url=https://library.semo.edu:4836/docview/222639538?accountid=38003

Walsh, J. A., Topol, E. J., & Steinhubl, S. R. (2014). Novel Wireless Devices for Cardiac Monitoring. Circulation, 130(7), 573-581. doi:10.1161/circulationaha.114.009024

Chronic Disease Prevention and Health Promotion. (2017, June 28). Retrieved April 25, 2018, from https://www.cdc.gov/chronicdisease/overview/index.htm

Heart and Stroke Statistics. (n.d.). Retrieved April 25, 2018, from https://www.heart.org/HEARTORG/General/Heart-and-Stroke-Association-Statistics_UCM_319064_SubHomePage.jsp

Overweight & Obesity. (2017, August 29). Retrieved April 25, 2018, from https://www.cdc.gov/obesity/childhood