HeartBeat Newsletter

DISEASE MANAGEMENT OF CHF BY MONITORING CO/CI

 

With a prevalence of 5.8 million in the US alone (2012), heart failure (HF) is a common syndrome associated with substantial morbidity, mortality, and health-care expenditures. Close to 1 million HF hospitalizations occur annually in the United States, with the majority of these resulting from worsening congestion in patients previously diagnosed with HF. An estimated 37.2 billion dollars is spent each year on HF in the United States. HF is the leading cause of hospitalization in individuals 65 years of age and older. One third of individuals who are hospitalized for HF either die or require readmission within 60 days. Earlier identification and treatment of congestion together with improved care coordination, management of co-morbid conditions, and enhanced patient self-management has been shown to help to prevent hospitalizations (readmission) in patients with CHF.

 

Such home monitoring extends from the promotion of self-care and home visitations, to telemedicine and remote monitoring of external or implantable devices. The incidence of HF increases with age. According to the Centers for Disease Control, among the U.S. residents who have HF, 70 percent are 60 years of age or older. It is estimated that in 2020, 16.5 percent will be in this age group which will lead to a significant increase in the prevalence of HF is expected in coming years. According to the Centers for Disease Control, more than 20 percent of men will develop HF within six years of having a heart attack. An even higher percentage (more than 40 percent) of women will suffer from HF within that period of time after having a heart attack. Together, the aging of the U.S. population and an improved medical outlook for heart attack victims account for the approximate threefold increase in the yearly incidence of HF that has been observed over the past 10 years.

These statistics emphasize the need to develop and implement more effective strategies to assess, monitor, and treat HF. Interventions geared towards identifying and monitoring sub-clinical congestion would be of value in the home management of chronic HF. Recent studies show that hospitals fail in decreasing the “30 days readmission” rate, and recently more than 2000 hospitals were penalized for that. Although Medicare is willing to employ companies for disease management of HF and DM, 5 out 8 companies that treated 240,000 patients, decided to discontinue their participation as they could not succeed to reduce costs, compared to the baseline before the program started.

It is assumed that a major factor in this failure of the hospitals and the disease management companies was the lack of suitable tools to continuously monitor the health state of the patients in real time and to administer suitable interventions, especially low-cost interventions of Lifestyle to complement medication and keep the patients at their homes with reasonable quality of life.

 

Existing Cardiac-Output devices

 

The gold standard for Cardiac-Output is an invasive procedure called Thermodilution (Swan-Ganz). There is also a non-invasive measurement using echocardiography and PC-MRI, but they are expensive, require skilled doctor and confined to the hospital. So is Edwards Lifescience Vigileo Flow-Trac that is based on invasive BP (A-Line).

The non-invasive devices based on Impedance Cardio-Graph (ICG) are estimating the changes in fluid in the chest using 6 to 8 electrodes. Systems like that are offered by Cheetah NICOM and PhysioFlow (see pictures below). Another device is offered by BMEye (that was recently acquired by Edwards LifeScience).

 

The main problem of the existing invasive and non-invasive devices mentioned above is that none of them is suitable for using by the patient at home and that their cost might be prohibitive for "Frugal" markets. The cost is many thousands and even tens of thousands dollars, they require skilled healthcare giver to place them on the patient and all of them need disposable sensors that are unacceptable for daily use at home. Also, all of them interfere with the daily life of the patient.

 

Another major problem that makes them not practical for large scale disease management is that all of them provide raw data that needs a skilled doctor interpretation. Just dumping continuous streams of data on the doctor is a useless strategy, as it does not have real economical advantage.

 

HeartBeat Solution

Our suggested solution is a combination of a mobile wearable device that monitors Cardiac Output (CO) / Cardiac Index (CI) and other hemodynamic parameters relevant to HF, and an Android Smart-phone App, that extracts Clinical significant data and helps with treatment as well as communicating with a Cloud server. The goal is to provide the tools for monitoring and disease management to optimize treatment and minimize cost in ER visits, hospitalization and readmission.

By using the above configuration of a wrist Pulse Oximeter with wireless connection to a mobile App we get a continuous PPG (PhotoPlethysmoGraph) signal that by our proprietry analysis can be employed to estimate continuously BP, CO/CI and other hemodynamic parameters.

 

In contrast with the all the other devices, this is the ONLY device in the world that is true wearable device that does not interfere with daily life, can be easily used by even the elderly and most sick patients at their homes, light weight (45 grams including rechargeable batteries), no extra cost of disposable and is low cost.

In contrast to all the other devices that required skilled placement of electrodes (ICG devices), or constant pressure and carrying and awakward pumping mechanism and wires (BMEye), the only thing needed here is wearing the watch! The low battery consumption and automatic turning off when sensor is removed makes it very simple even for the most unsophisticated, old and sick patient.

 

At this point we offer existing watch with existing thumb sensor, but we prepare also a version when the reflective sensor is inside the watch.

 

Cardiac Output (CO) / Cardiac Index (CI)

 

HF results when the heart cannot efficiently supply the body with blood. I.e. the Cardiac Output (or better Cardiac Index where CO is normalized by the Body Surface Area) is not sufficient to meet the demands of the body. Therefore monitoring of CO/CI  is the best and most direct way to measure and monitor HF.  

Cardiac Index divides the CO by the estimated Body Surface Area (BSA) and by that normalizes the CO value to various body sizes so therefore it sis a better estimator of how much blood is supplied to the tissues. Another important issue is the CO/CI during resting vs. CO/CI during efforts. Many times the CO/CI might look sufficient during resting, but the real test is during effort, when the blood supply is in higher demand. One of the biggest advantages of our watch is that it can follow the patient during her/his daily life activities.

Diastolic and systolic are the two basic types of HF:

Diastolic HF happens when the heart cannot properly fill with blood.

Systolic HF, the more common of the two, occurs when the heart does not efficiently pump blood from the ventricles to the body.

The result of either type of HF is decreased CO. Less blood is pumped from the heart to the body. Decreased CO also can lead to decreased blood pressure.

Many things can lead to HF. Systolic HF is commonly caused by a heart attack and/or persistent high blood pressure. Diastolic HF may be a result of systolic HF, dysfunctional heart valves, or a diseased heart lining. Hypertension is one of the most common causes. Other major risk factors are diabetes mellitus, high cholesterol, obesity, and smoking. The continuous measurement of Hemodynamics by our system will assist the doctor in managing the treatment of the HF patient, keeping her/him away from the hospital.

 

There are several guidelines for the diagnosis and management of Heart Failure.  Frequent or continuous monitoring of Cardiac output and other vital signs may help to prevent or postpone re-admission better adherence and better dosing of medication and also helps to provide rapid feed back on lifestyle measures like exercise, diet and supplementation of minerals and vitamins.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The presentation of the processed data on the Smart-phone display is explained in more details below. The main objective is to present all relevant data in a graphic concise way, so the doctor can in a short glance get the picture about the patient status, trend over time and relationship among all components, in order to save the doctor’s time and promote more efficient treatment.

The S/W avoids doing medical interpretation, as this is the doctor’s responsibility.

 

There are many different ways to categorize HF, including:

• the side of the heart involved (left HF versus right HF). Right HF compromises pulmonary flow to the lungs. Left HF compromises aortic flow to the body and brain. Mixed presentations are common; left HF often leads to right HF in the longer term.

• whether the abnormality is due to insufficient contraction (systolic dysfunction), or due to insufficient relaxation of the heart (diastolic dysfunction), or to both.

• whether the problem is primarily increased venous back pressure (preload), or failure to supply adequate arterial perfusion (afterload).

• whether the abnormality is due to low CO with high systemic vascular resistance or high CO with low vascular resistance (low-output HF vs. high-output HF).

• the degree of functional impairment conferred by the abnormality (as reflected in the New York Heart Association Functional Classification[9])

• the degree of coexisting illness: i.e. HF/systemic hypertension, HF/pulmonary hypertension, HF/diabetes, HF/renal failure, etc.

All the above are the doctor’s decisions and once the results are sent to the Cloud server, the treating doctor is encourage to enter her/his diagnosis and treatment to the patient’s file.

This information will be used as well (without the patient’s name (that will be only in the patient smart-phone), to generate statistical inference over all HF population.

The Cloud server and inference engine are out of the scope of this current project, and will be developed only if there is a demand for that. It will be based on the author Fuzzy Machine Learning algorithm that has been developed for another project. This engine combines data, diagnosis and treatment of many patients to figure out the relative importance and contribution of each component to the total health score of the patient, leading to accumulated knowledge for optimizing diagnostics and treatment.

However, these statistical inferences will be for recommendation and information purposes only, and will not affect the diagnosis or treatment decisions.

 

Data Display

 

The data-display plays a major role in the usefulness of the Personal Server to the Monitoring and Disease management providers. As stated above, the representation of the data and its graphical presentation, as well as specific calculations, are just to save time for the doctor and do not involve any medical interpretation or decisions. The diagnosis and treatment decisions are the doctor’s responsibility and the suggested S/W does not claim any diagnosis or treatment abilities.

 

Prior to activating the GUI, an input table should contain the following info:

 

1. ID of patient – integer number

2. Gender

3. Age

4. Height (in meters) – e.g. 1.73

5. Weight (in Kg)  - e.g. 66 Kg

6. BSA (Body Surface Area) – computed from 4 and 5:

 

The Mosteller¹ formula

BSA (m²) = ( [Height(cm) x Weight(kg) ]/ 3600 )½         e.g. BSA = SQRT( (cm*kg)/3600 ) or in inches and pounds:      BSA (m²) = ( [Height(in) x Weight(lbs) ]/ 3131 )½

 

7. List of parameters to be displayed (as determined by the doctor)

e.g. CO (Cardiac Output), CI = CO/BSA, SpO2, BP, Activity (calories or kW – computed from 3D accelerometer), Respiration (Res per minute), temperature, ECG, etc..

8 . Series of numbers expressing relative importance for each chosen parameter

(e.g. CO – 0.3; BP – 0.15, …)

9. For each chosen parameter – range of normal values. E.g. CO – 4-6 L/min: BP – 120 -100 Sys; 60-80 Diastolic ‘ etc..

10. Alarm thresholds for each chosen parameter

 

This table has to be filled only once, and updated if necessary. It can be populated either by the treating Physician or from known norms. Some values can be sent from devices like Bluetooth Weight Scale, or computed from other enteries (e,g, BSA)

 

This table can be changed by the Physician at any time, depending on the progress of the disease.The display has top buttons and 2 panes: Pie-Chart and Time graph

 

 

As we mentioned above, in order to keep the regulatory approval to be as simple as possible, we avoid any actions or claims that can be construed as performing diagnostics or affecting treatment, We leave all of that to the doctor and confine our contribution to better presentation of data acquired by regulatory approved devices like Pulse Oximeter.

 

We believe that the device as is, will be a great contribution and will turn to be a “must have” device for cost efficient HF disease management , that can be used both in the hospital and at home.

 

However, we are preparing the next generations that might include integration of additional signals like ECG, Phonocardiography for monitoring heart sounds and other signals.

 

Also, we are exploring new sensors that could be integrated inside the watch and provide additional information.

We also explore the Cloud server side to provide accumulated statistical inference from all users of the device, taking advantage on the connectivity aspect of the smart-phone. As mentioned above, these are all future directions and are not included in this project.

 

References

Abramowitz H B, Queral L A, Finn W R, Nora P F Jr, Peterson L K, Bergan J J and Yao J S 1979 The use of photoplethysmography in the assessment of venous insufficiency: a comparison to venous pressure measurements Surgery 86 434–41

Agache P G and Dupond A S 1994 Recent advances in non-invasive assessment of human skin blood flow Acta. Derm. Venereol. Suppl. (Stockholm) 185 47–51 R30 Topical Review

Ahmed A K, Harness J B and Mearns A J 1982 Respiratory control of heart rate Eur. J. Appl. Physiol. 50 95–104

Aldrich T K,Moosikasuwan M, Shah S D and Deshpande K S 2002 Length-normalized pulse photoplethysmography: a noninvasive method to measure blood hemoglobin Ann. Biomed. Eng. 30 1291–8

Allen J 2002 The measurement and analysis of multi-site photoplethysmographic pulse waveforms in health and arterial disease PhD Thesis Newcastle University

Allen J, Frame J R and MurrayA2002 Microvascular blood flowand skin temperature changes in the fingers following a deep inspiratory gasp Physiol. Meas. 23 365–73

Allen J and Murray A 1993 Development of a neural network screening aid for diagnosing lower limb peripheral vascular disease from photoelectric plethysmography pulse waveforms Physiol. Meas. 14 13–22

Allen J and Murray A 1995 Prospective assessment of an artificial neural network for the detection of peripheral vascular disease from lower limb pulse waveforms Physiol. Meas. 16 29–38

Allen J and Murray A 1996 Comparison of three arterial pulse waveform classification techniques J. Med. Eng. Tech. 20 109–14

Allen J and Murray A 1999 Modelling the relationship between peripheral blood pressure and blood volume pulses using linear and neural network system identification techniques Physiol. Meas. 20 287–301

Allen J and Murray A 2000a Similarity in bilateral photoplethysmographic peripheral pulse wave characteristics at the ears, thumbs and toes Physiol. Meas. 21 369–77

Allen J and Murray A 2000b Variability of photoplethysmography peripheral pulse measurements at the ears, thumbs, and toes IEEE. Proc. Sci. Meas. Technol. 147 403–7

Allen J and Murray A 2002 Age-related changes in peripheral pulse timing characteristics at the ears, fingers and toes, J. Hum. Hypertens. 16 711–7

Allen J and Murray A 2003 Age-related changes in peripheral pulse shape characteristics at various body sites Physiol.Meas. 24 297–307

Allen J and Murray A 2004 Effects of filtering on multi-site photoplethysmography pulse waveform characteristics IEEE Comput. Cardiol. 31 485–8

Allen J, Oates C P, Lees T A and Murray A 2005 Photoplethysmography detection of lower limb peripheral arterial occlusive disease: a comparison of pulse timing, amplitude and shape characteristics Physiol. Meas. 26 811–21

Allen J, Overbeck K, Stansby G and Murray A 2006 Photoplethysmography assessments in cardiovascular disease Meas. Control 39 80–3

Almond N E and Cooke E D 1989 Observations on the photoplethysmograph pulse derived from a laser Doppler flowmeter Clin. Phys. Physiol. Meas. 10 137–45

Anderson R R and Parrish J A 1981 The optics of human skin J. Invest. Dermatol. 77 13–9

Aoyagi T, Kiahi M, Yamaguchi K and Watanabe S 1974 Improvement of the earpiece oximeter Abstracts of the 13th Annual Meeting of the Japanese Society of Medical Electronics and Biological Engineering 90–1

Aoyagi T and Miyasaka K 2002 Pulse oximetry: its invention, contribution to medicine, and future tasks Anesth. Analg. 94 (Suppl. 1) 51–3

Avnon Y, Nitzan M, Sprecher E, Rogowski Z and Yarnitsky D 2003 Different patterns of parasympathetic activation in uni- and bilateral migraineurs Brain 126 1660–70

Avnon Y, Nitzan M, Sprecher E, Rogowski Z and Yarnitsky D 2004 Autonomic asymmetry in migraine: augmented parasympathetic activation in left unilateral migraineurs Brain 127 2099–108

Avolio A P, Chen S G, Wang R P, Zhang C L, Li M F and O’Rourke M F 1983 Effects of aging on changing arterial compliance and left ventricular load in a northern Chinese urban community Circulation 68 50–8

Azabji Kenfack M, Lador F, Licker M, Moia C, Tam E, Capelli C, Morel D and Ferretti G 2004 Cardiac output by Modelflow method from intra-arterial and fingertip pulse pressure profiles Clin. Sci. 106 365–9

Babchenko A, Davidson E, Adler D, Ginosar Y, Kurz V and Nitzan M 2000 Increase in pulse transit time to the foot after epidural anaesthesia treatment Med. Biol. Eng. Comput. 38 674–9

Barnes R W 1991 Noninvasive diagnostic assessment of peripheral vascular disease Circulation 83 (2 suppl.) I20–7

Barnes R W, Clayton J M, Bone G E, Slaymaker E E and Reinertson J 1977a Supraorbital photoplethysmography. Simple, accurate screening for carotid occlusive disease J. Surg. Res. 22 319–27

Barnes R W, Garrett W V, Slaymaker E E and Reinertson J E 1977b Doppler ultrasound and supraorbital photoplethysmography for noninvasive screening of carotid occlusive disease Am. J. Surg. 134 183–6

Barron S A, Rogowski Z, Kanter Y and Hemli J 1993 DC photoplethysmography in the evaluation of sympathetic vasomotor responses Clin. Physiol. 13 561–72

Belcaro G et al 1998 Noninvasive investigations in vascular disease Angiology 49 673–706