AONA
`
`20040059236A1
`
`as) United States
`a2) Patent Application Publication o) Pub. No.: US 2004/0059236 Al
`(43) Pub. Date: Mar. 25, 2004
`
`Margulies etal.
`
`(54) METHOD AND APPARATUS FOR
`MONITORING THE AUTONOMIC NERVOUS
`SYSTEM
`
`(76)
`
`Inventors: Lyle Aaron Margulies, Seattle, WA
`(US); David B. Harrell, Mukilteo, WA
`(US); Michael Allen Riggins, Seattle,
`WA (US)
`
`Correspondence Address:
`GARRISON ASSOCIATES
`2001 SIXTH AVENUE
`SUITE 3300
`SEATTLE, WA 981212522
`
`(21) Appl. No.:
`
`10/666,121
`
`(22)
`
`Filed:
`
`Sep. 19, 2003
`
`Related U.S. Application Data
`
`(60)
`
`Provisional application No. 60/412,310,filed on Sep.
`20, 2002.
`
`Publication Classification
`
`Lint. C17 ecsccsscccssscssesesescceenssseeevnnseeeevnseee A61B 5/02
`(SL)
`(52) USh@ly scnsseennsacciauneetacian 600/500
`
`(57)
`
`ABSTRACT
`
`An apparatus and method for detection and monitoring of
`autonomic nervous system (ANS) activity in humans,pri-
`marily in the field of sleep research. The present invention
`discloses a portable, simple, and cost-effective electronic
`device containing hardware and software that permits real-
`time monitoring of a pulsatile blood volume waveform
`obtained through use of a photoplethysmographic (optical
`volume detecting) probe, thereby allowing signal condition-
`ing, waveformslope analysis, display, recording, and output
`ofpulse transitional slope data representative of activity in
`the ANS.
`
`LIGHT SOURCE
`
`
`
`PHOTODETECTOR
`
`001
`
`U.S. Patent No.
`
`Apple Inc.
`APL1034
`8,923,941
`
`Apple Inc.
`APL1034
`U.S. Patent No. 8,923,941
`
`001
`
`

`

`Patent Application Publication Mar. 25,2004 Sheet 1 of 7
`
`US 2004/0059236 Al
`
`FIG.1
`
`LIGHT SOURCE
`
`
`
`PHOTODETECTOR
`
`
`
`
`
`
`ee Sseeee 8 Os
`
`
`
`ABSORPTION
`
`
`
`
`VARIABLE ABSORPTION
`DUE TO PULSE-ADDED
`VOLUME OF ARTERIAL
`BLOOD
`ABSORPTION DUE
`TO ARTERIAL BLOOD
`
`ABSORPTION DUE
`TO VENOUS BLOOD
`
`ABSORPTION.
`DUE TO TISSUE
`
`TIME
`
`FIG.2
`
`002
`
`002
`
`

`

`Patent Application Publication Mar. 25,2004 Sheet 2 of 7
`
`US 2004/0059236 Al
`
`FIG.3
`
`gg—____. PEAK TOP
`
`SLOPE END
`
`PEAK BOTTOM
`
`
`
`dP/dt
`
`PERIPHERAL PULSE WAVEFORM
`
`003
`
`003
`
`

`

`ah"
`
`ool
`
`US 2004/0059236 Al
`
`Patent Application Publication Mar. 25,2004 Sheet 3 of 7
`
`dr’OASINLOVTOULNOD
`drOldyooyoo“eJoI~OLNLOVTOULNODVY1+y¥Vv“VYNy993
`VvOldSINLOVTOYLNOOD
`
`004
`
`004
`
`

`

`
`
`68S818242€269G919LSES6hShIbLE€€62GZ12LbEt
`
`6 |
`
`
`
`$L1V39LYVSH
`
`
`
`Patent Application Publication Mar. 25,2004 Sheet 4 of 7
`
`US 2004/0059236 Al
`
`
`
`ddO1S
`
`OllvY
`
`005
`
`005
`
`

`

`
`
`
`
`
`
`
`
`LLBEBHIELCEEBLES962LELZ60SZ1872ESOZSZBlLESLGOELLhLL€16$89LSP62%|
`
`
`
`
`
`
`
`S1V34LYV3HqZ°Old
`
`Patent Application Publication
`
`Mar.25, 2004 Sheet 5 of 7
`
`US 2004/0059236 Al
`
`Vi0iId4
`
`LYVSH@hweMAWAM:eetWNTseyyTeyoeeeee
`
`SLV38
`
`006
`
`006
`
`
`
`

`

`Patent Application Publication
`
`Mar. 25, 2004 Sheet 6 of 7
`
`US 2004/0059236 Al
`
`WIY¥3Ss
`
`LYOd
`
`SOVAYNSILNI
`
`YslNdWoo918
`
`LL
`
`WIMAS
`
`O/l
`
`018
`
`
`
`sngidsONVAYS1LVEAS
`
`YAMOdAC’
`
`peesi
`
`AlddNs
`
`3J5VyOSl8‘aNQOVIGSWILINW
`
`
`
`NOLINGdOLS/LYYLSyasn
`
`NOLINGLINSNYHLgTOuINOD
`
`
`YSATIOULNOODNIV\AONANDIYS
`
`98wats28
`
`-ONDINTOYLNODS8MO1360ud
`
`LNdLno
`
`
`
`HdWY9A10dWas
`
`LYOd
`
`68iI88AV1gSi0
`
`
`
`TWIYaS381d
`
`HdVu9YvE
`
`LYuOd
`
`HdVu9Uva
`
`€Lg=
`
`
` ne—|ELndino=HdVY9ATOdeel
`
`
`
`OllWe34078
`
`yYOTOOId
`
`qa
`
`ce18AONANOSYS
`
`yal
`
`HOIH8€9
`
`3g0ud
`
`30V4Y3INI
`
`007
`
`007
`
`
`
`

`

`Patent Application Publication Mar. 25,2004 Sheet 7 of 7
`
`US 2004/0059236 Al
`
`DISPOSING A
`PHOTO-PLETHYSMOGRAPHIC PROBE
`PROXIMAL TO A SINGLE BODY PART
`
`
`
`
`
`DERIVING A CONTINUOUS PULSATILE
`BLOOD VOLUME WAVEFORMAS A
`FUNCTION OF PULSE AMPLITUDE AND
`TIME
`
`FIG.9
`
`PROVIDING AN
`
`
`
`OFINESRASLTION
`REPRESENTATIVE
`OF SLOPE VALUES
`
`DEFINING A TIME INTERVAL FOR
`CALCULATION OF A SLOPE OF THE
`PULSATILE BLOOD VOLUME WAVEFORM
`
`TIME INTERVAL
`
`PERFORMING CONTINUOUS
`CALCULATION OF THE SLOPE OF THE
`RISING SEGMENT OF EACH BLOOD
`VOLUME WAVEFORM OVER DEFINED
`
`PROCESSING INPUT DATA TO DIVIDE
`PEAK AMPLITUDE VALUES BY A GIVEN
`TIME CONSTANT
`
`ELIMINATING FROM FURTHER
`CALCULATION SLOPE VALUES OF LESS
`THAN ONE
`
`SIGNAL PROCESSING, CONDITIONING,
`AND ARTIFACT REJECTION
`
`AMPLIFYING AND FILTERING SLOPE
`roa = m on
`
`A<,
`
`
`
`PROVIDING DATA
`
`
`OUTPUT
`REPRESENTATIVE
`OF SLOPE VALUES
`
`
`FOR USE BY
`
`
`OTHER DEVICES
`
`
`
`008
`
`STORING
`
`
`
`ELECTRON IC DATA
`REPRESENTATIVE
`OF SLOPE VALUES
`
`008
`
`

`

`US 2004/0059236 Al
`
`Mar. 25, 2004
`
`METHOD AND APPARATUS FOR MONITORING
`THE AUTONOMIC NERVOUS SYSTEM
`
`CROSS-REFERENCE TO RELATED
`APPLICATION
`
`(0001] This application claims benefit of United States
`Provisional Application Serial No. 60/412,310 entitled
`Method and Apparatus for Monitoring the Autonomous
`Nervous System, filed Sep. 20, 2002.
`
`TECHNICALFIELD
`
`(0002] This invention relates to medical devices, and more
`particularly to physiological monitoring methods and
`devices used for detection of autonomic nervous system
`(ANS) activity in the field of sleep research. The present
`invention discloses a portable, simple, and cost-effective
`electronic sleep diagnostic device containing hardware and
`software that permits recording and signal conditioning of a
`pulsatile blood volume waveform obtained through use of a
`photoplethysmographic (optical volume detecting) probe,
`thereby allowing analysis pulse transitional slope data that is
`representative of activity in the autonomic nervous system
`(ANS).
`
`BACKGROUND OF THE INVENTION
`
`risk is directly linked to sleep
`[0003] Cardiovascular
`related breathing disorders (SRBD). The number of U.S.
`laboratories that study sleep, roughly 2,792,
`is incredibly
`low when compared to the number of Americans estimated
`to have a chronic SRBD,just over 40 million, The average
`number of beds per lab is 3.6 bringing the total number of
`beds in which to do a sleep study to roughly 10,000. This
`meansthat to test all 40 million Americans, there would be
`4,000patients that would be seen perbed. If sleep tests were
`run 365 days per year, the result is an astounding 11 years
`of conclusive tests needed to be run to test
`the current
`population ofindividuals suffering form SRBD. The length
`of time increases as one considers the actual number of days
`per year sleep labs actually test patients, plus the amount of
`tests that need to be re-run due to inconclusive testing, plus
`the numberofpatients that continually need to be retested to
`see if their treatment is functioning properly. Given this
`scenario, it is no shock that wait times for patients to be
`scheduledfor a sleep test can typically range from six weeks
`to six months. The problem will only increase, as “it
`is
`estimated that nearly 80 million Americans will have a sleep
`problem by the year 2010 and 100 million will have one by
`the year 2050.” Clearly then, the problem with wait time for
`testing should be addressed immediately to relieve pent up
`demand.
`
`testing sleep
`[0004] The current “gold standard” for
`related breathing disorders is full polysomnography. Full
`polysomnography is, however, quite
`labor
`intensive,
`requires considerable instrumentation andis therefore rather
`expensive to conduct. As a result, many sleep laboratories
`have found it difficult to keep up with the demand for this
`test, and a long waiting list becomes the norm. Given that
`obstructive sleep apnea (OSA) is quite prevalent, leads to
`serious complications and that treatment options exist, it is
`important
`that individuals suffering from the disease are
`identified.
`
`(0005] The need to study the ANS has been realized in
`academia for a considerable time. It is knownin the field of
`
`microneurography that rapid-eye movement (REM) sleep is
`associated with profound sympathetic activity. It has also
`been found that arousals from non-rapid-eye movement
`(NREM) elicits K complexes that are associated with sym-
`pathetic activity. The sympathetic division of the ANS
`prepares a body for movement. Arousals require movement
`and hence an arousal requires sympathetic activation.
`
`[0006] Generally, patients with OSA, a type of SRBD,
`have extremely disrupted sleep and terribly high daytime
`somnolence. Obstructive sleep apnea events are always
`accompanied by an acute rise in systolic blood pressure
`(rises in systolic blood pressure are associated with sympa-
`thetic activation), even when the usual EEG criteria for
`arousals are not met (a recognizable cortical electroencepha-
`lographic arousal). The duration of the apnea ofindividuals
`that demonstrate EEG arousal and those that do not meet the
`
`usual criteria for defining an arousal have been found to be
`identical. The pleural pressure peak, at the end of apnea,is
`identical between the two types of arousals, as are the EEG
`frequencies. These findings suggest
`that monitoring the
`cardiac changes of sleep is a more accurate measurement.
`[0007]
`It has been demonstrated that apneic episodes
`result in progressive increases in sympathetic nerve activity.
`The increases are most marked toward the end of the apnea,
`when a patient moves. These findings are exactly what is
`excepted of sympathetic activation and its relationship to
`arousals in patients with SRBD.
`[0008] Because cardiovascular control during sleepis pri-
`marily dictated by brain states that produce profoundvaria-
`tion in ANSactivity, many studies have been conducted to
`monitor the ANS. Since the data shows clearly that moni-
`toring the ANS or cardiac changes in sleep yields more
`accurate data defining an arousal
`in sleep,
`it
`is clear that
`diagnostic studies must include ANS or cardiac monitoring.
`
`It has been shown that in transitions from NREM
`[0009]
`to REM sleep, heart rate accelerations precede the EEG
`arousals marking the onset of REM. Therefore, not only
`does monitoring ANS activity give the clinician a possibly
`more accurate study, but also changes in ANS activity
`precede that information being observed via the EEG elec-
`trodes.
`
`‘There are two existing technologies that attempt to
`[0010]
`monitor the ANS, namely pulse transit
`time (PTT) and
`peripheral arterial tonometry (PAT). Neither PTT nor PAT
`can lay claim to monitoring the ANS without adding addi-
`tional sensors. PTT requires the use of ECG electrodes that
`may be difficult for a patient
`to self-apply due to skin
`cleaning and shaving requirements. PAT requires a very
`costly gauntlet-type device with a single-use finger pressure
`cuff. Also, the addition of extra sensors adds to noise artifact
`and difficulty in patient use. It is therefore an object of the
`present invention to provide an improvement over existing
`PTT and PAT technology through a more economical and
`more easily used device without need of additional sensors.
`
`[0011] Several disclosures have been madein thepriorart
`that teach methods and devices for diagnosis and monitoring
`of sleep breathing disorders using physiological data
`obtained from pulse oximetry-derived waveforms.
`
`[0012] U.S. Pat. No. 5,398,682 to Lynn (Mar. 21, 1995)
`discloses a method and apparatus for the diagnosis of sleep
`apnea utilizing a single interface with a human bodypart.
`
`009
`
`009
`
`

`

`US 2004/0059236 Al
`
`Mar. 25, 2004
`
`Morespecifically, a device is disclosed for diagnosing sleep
`apnea by identifying the desaturation and resaturation events
`in oxygen saturation of a patient’s blood. The slope of the
`events is determined and compared against various infor-
`mation to determine sleep apnea.
`
`[0013] U.S. Pat. No. 6,363,270 B1 to Colla, et al. (Mar. 26,
`2002) discloses a method and apparatus for monitoring the
`occurrence of apneic and hypopneic arousals utilizing sen-
`sors placed on a patient to obtain signals representative of at
`least two physiological variables, including blood oxygen
`concentration, and providing a means for recording the
`occurrence of arousals. Obtained signals pass through con-
`dilioning circuitry and then processing circuitry, where
`correlation analysis is performed. A coincident change in at
`least
`two of the processed signals are indicative of the
`occurrence of an arousalthat in turn indicates an apneic or
`hypopneic episode has occurred. A patient
`thus can be
`diagnosed as suffering conditions such as obstructive sleep
`apnea.
`
`[0014] U.S. Pat. No. 6,529,752 B2 to Krausman and Allen
`(Mar. 4, 2003) discloses a method and apparatus for count-
`ing the number of sleep disordered breathing events expe-
`rienced by a subject within a specified time period. Such a
`counter comprises: (1) an oxygen saturation level sensor for
`location at a prescribedsite on the subject, (2) an oximetry
`conditioning and control module that controls the operation
`of the sensor and converts its output data to oxygen satu-
`ration level data, (3) a miniature monitoring unit having a
`microprocessor, a memory device, a timer for use in time-
`stamping data, a display means and a recall switch, and (4)
`firmware for the unit
`that directs: (i) the sampling and
`temporary storage of the oxygen saturation level data, (il)
`the unit to analyze using a specified method the temporarily
`stored data to identify and count
`the occurrence of the
`subject's disordered breathing events, and to store the time
`of occurrence of each of these events, and (ii) the display
`means to display specified information pertaining to the
`counts in response to the actuation of the recall switch.
`
`[0015] U.S. Pat. No. 6,580,944 B1 to Katz, et al. (Jun. 17,
`2003) discloses a method and apparatus for identifying the
`uumingof the onset of and duration of an event characteristic
`ofsleep breathing disorder while a patient is awake. Chaotic
`processing techniques analyze data concerning a cardiores-
`piratory function, such as oxygen saturation and nasal air
`flow. Excursions ofthe resulting signal beyond a threshold
`provide markers for delivering the averagerepetition rate for
`such events that is useful in the diagnosis of obstructed sleep
`apnea and other respiratory dysfunctions.
`[0016] The above references all make use of oxygen
`saturation data obtained through pulse oximetry to deter-
`mine arousals and/or sleep breathing disorders. Each nec-
`essarily requires additional analysis and calculation of blood
`oxygen concentrations in order to render information useful
`specifically in the diagnosis and monitoring ofsleep breath-
`ing disorders. It is therefore another object of the present
`invention to provide a more simplified method of obtaining
`and analyzing physiological data that accurately represents
`ANSactivity.
`
`BRIEF SUMMARY OF THE INVENTION
`
`Itis an object of the present invention to overcome
`(0017]
`one or more of the problems with the prior art.
`In one
`
`invention provides a
`the present
`preferred embodiment
`method and apparatus for improved monitoring of ANS
`activity using a single patient sensor.
`[0018] A variety of breathing disturbances may occur
`during sleep,
`including snoring, hypoventilation, apnea,
`increased upper-airway resistance, and asthma related con-
`ditions. This project proposes developmentofa novel device
`that can noninvasively and accurately detect frequent brief
`micro arousals that are not well identified by conventional
`airflow, respiratory effort, pulse oximetry and EEG methods.
`These subcortical events result from increased respiratory
`effort and cause disruption of nocturnal sleep, leading to
`excessive daytime somnolence.
`
`[0019] Since microarousals have been associated with
`changes in autonomic system outflow, this invention pro-
`vides for a small, portable device that analyzes the shape of
`the arterial finger pulse, thereby detecting on a beat by beat
`basis changes in vascular
`tone directly attributable to
`microarousals. The present invention uses a photoplethys-
`mographically derived arterial blood volume waveform for
`monitoring changes in peripheral arterial vascular tone, in
`conjunction with A/D converters and a microcontroller for
`analyzing the morphologyofthe pulsatile signal.
`
`[0020] The method of the present invention provides for
`detection of microarousals that compares favorably with
`detection by pulse transit time (PTT) devices, EEG analysis,
`ECGanalysis, esophagal pressure (Pes) or some combina-
`tion of these methods. Although PTT and peripheral arterial
`tonometry (PAT) have both been receiving muchattention as
`techniques for detecting changes in the ANS during sleep
`studies, PAT is relatively expensive and PTT has implemen-
`tation problems caused by motion artifact.
`[0021]
`It
`is a further object of the present invention to
`provide an apparatusthat utilizes transmittedlight intensity
`from an existing FDA approved pulse oximeter probe sothat
`no additional device is attached to the patient. Valuable
`diagnostic information can then be extracted through elec-
`tronic processing of this existing data.
`[0022] Normalization is a method to correct for the pho-
`toplethysmographic pulse signal morphological changes
`based on finger position (as opposedto actual changes of
`autonomic activity.) PTT and PATlack a means for signal
`normalization and therefor cannot correctfor finger position
`changes. Normalization provides
`immunity to artifact
`caused by both elevation changes of the finger probe, and
`changes in blood flow due to arterial compression during
`patient positional changes. It is therefor another objectof the
`present invention to provide a means of normalization in
`order to ensure appropriate artifact suppression.
`
`[0023] Since pulse oximeters use an alternating flashing of
`two different wavelength LEDs,
`the present
`invention is
`intended to synchronize with the desired LED in order to
`examine the transmitted intensity due to a single wave-
`length. Alternatively, certain models of oximeter OEM mod-
`ules provide an analog or digital output that can be utilized
`directly by the present invention.
`
`[0024] Another objective is to provide algorithms for
`slope detection, peak to peak height, and normalization may
`be performedeither with firmware within the present inven-
`tion apparatus, or by software after the data is downloaded
`into a polysomnograph or other data processing device.
`
`010
`
`010
`
`

`

`US 2004/0059236 Al
`
`Mar. 25, 2004
`
`It is a further objective of the present invention to
`[0025]
`provide a meansofdata storage and transfer, and to provide
`a method of displaying the observed changes in slope.
`Alternative embodiments display these changes as a wave-
`form, light bars, and/or numerical information.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`1 shows a schematic representation of a
`[0026] FIG.
`typical pulse oximeter sensing configuration on a finger.
`
`[0027] FIG. 2 shows a graphic representation of the
`components of vascular Ussue that contribute to light
`absorption plotted as absorption versus time.
`
`a detailed waveform of pressure versus time. This continu-
`ous pulse wave tracing contains precise waveshape, fre-
`quency, and inflection information easily discernable by the
`human eye that
`is not available from only systolic and
`diastolic pressure numerics. The progression from pressure
`transducers to photoplethysmography allows detection of
`the pulse wave at sites not easily palpated, including the
`finger and earlobe. Photoplethysmography detects the
`changes in the amount of light absorbed by hemoglobin,
`which corresponds to changes in blood volume. Changes in
`amplitude of the photoplethysmographic wave have been
`used to evaluate arterial compliance, but the wave contour
`itself was not used, as is disclosed by the present invention.
`
`(0028] FIG. 3 shows a graphic representation of a single
`peripheral pulse waveform plotted as volume versus time.
`
`Plethysmography is the measurement of volume
`(0038]
`changesoftissue or an organ. Photoplethysmography mea-
`[0029] FIG. 4 shows comparative physiological wave-
`sures blood volume changesinatissue using the fractional
`forms following administration of vasoactive agents.
`change in light
`transmission. One of the most common
`applicationsof this technology is the noninvasive measure-
`[0030] FIG. 5 shows a second derivative waveform con-
`ment of the oxygen saturation of the hemoglobin in red
`sisting of a, b, c and d waves in systole, and an e wave in
`diastole.
`blood cells through a technique called pulse oximetry. FIG.
`1 showsa typical pulse oximeter sensing configuration on a
`finger. Typically, two different wavelengths oflight (e.g. 660
`and 805 nm) are applied to one side of a finger and the
`received intensity is detected on the opposite side after
`experiencing some absorption by the intervening vascular
`tissues. The amount of absorption (and conversely transmis-
`sion) is a function of the thickness,color, and structure of the
`skin, tissue, bone, blood, and other tissues that
`the light
`traverses.
`
`[0031] FIG. 6 showsa graphic representation of changes
`in Normalized Slope plotted as slope ratio versus heart beats
`while subject performs Valsalva maneuver.
`
`[0032] FIG. 7 shows a sleep stage hypnogram ofan hour
`and a quarter sleep study.
`[0033] FIG. 8 shows a block diagram of the present
`invention apparatus.
`
`[0034] FIG. 9 shows a block diagram of the present
`invention method.
`
`DETAILED DESCRIPTION OF THE
`INVENTION
`
`(0035] A variety of breathing disturbances may occur
`during sleep,
`including snoring, hypoventilation, apnea,
`increased upper-airway resistance, and asthma related con-
`ditions. The present invention discloses a method and appa-
`ratus that can noninvasively and accurately detect frequent
`brief “microarousals” (small amplitude subcortical distur-
`bancesthat disrupt normalsleep) that are not well identified
`by conventional airflow, respiratory effort, pulse oximetry
`and EEG methods. These subcortical events result from
`increased respiratory effort and cause disruption of nocturnal
`sleep, leading to excessive daytime somnolence.
`
`[0036] Microarousals can be detected using data obtained
`from the absorbanceofvisible or infrared light in a finger or
`other body part of a patient, and by analyzing changes in the
`obtained peripheral blood volume waveform that are indica-
`tive of microarousals. Specifically, sufficient information is
`contained in slope variations of the rising edge of the
`pulsatile blood volume waveform to allow analysis of
`changes in the autonomic nervous system (ANS). This
`technology is herein referred to as pulse transitional slope
`(PTS). Both ANS and hemodynamic responses occur during
`obstructive sleep apnea and are influenced by apnea, hypop-
`nea, hypercapnea, and arousal.
`
`(0037] Analysis of the noninvasive blood pressure pulse
`wave has been shownto be useful for evaluation of vascular
`load and aging. Pressure transducers located at a palpable
`artery, such as the carotid, femoral, or radial artery provided
`
`invention is specifically directed to
`[0039] The present
`alpha andrenergic receptor sites,
`the activation of these
`receptors at certain locations on the body resulting in
`physiological responses such as peripheral vascular resis-
`tance, mydriasis, and contraction of pilomotor muscles,
`which are representative of sympathetic nervous system
`activity. The preferred locations generally includethefingers
`and the big toe (other sites are under investigation), due to
`a desirable lack of beta or parasympathetic receptors at those
`locations on the body.
`
`[0040] The transmitting light comes from light emitting
`diodes (LEDs), typically in the visible red and the invisible
`infrared (IR) spectrums. The optical receiver may be a
`photodiode, photoresistor, or solar cell. By using two dif-
`ferent wavelengths, each with different absorbance charac-
`teristics in oxygenated and deoxygenated blood, the inten-
`sity ratio between the two received signals can be analyzed,
`and not just the intensity. Therefore the attenuating tissues
`mentioned earlier do not affect the ratio of the intensities,
`which via a look-up table can determine the oxygen satu-
`ration percent in the finger vasculature.
`
`[0041] FIG. 2 shows the components of vasculature tissue
`that contribute to light absorption. The static or de compo-
`nentofthe received optical signal represents light absorption
`by the tissue, venous blood, pigments and other structures.
`The present invention is concerned with the ac, or pulsatile
`component because the focus is on examining the wave
`shape ofthe systolic portion of the blood volume waveform.
`Electronically, the de component is removed with a simple
`resistor-capacitor high pass circuit
`that has a -3 dB fre-
`quency of around one Hertz.
`
`011
`
`011
`
`

`

`US 2004/0059236 Al
`
`Mar. 25, 2004
`
`[0042] The amountoflight passing through the finger is
`called transmittance, T, and is defined by:
`T=i/lo
`
`[0053]
`
`NTG=Nitroglycerin
`
`(0054] PTG=Photoplethysmography
`
`[0043] where Io is the intensity ofthe incidentlight and I
`is the intensity of the transmitted light.
`
`[0044] The amount of light of a specified wavelength
`absorbed by a substanceis directly proportional to both the
`length ofthe light path and the concentration of the material
`within the light path. The absorbance, A, is defined as the
`negative logarithm of the transmittance, or:
`A=-log T=-log //lo=aCh
`
`[0045] where a is a constant called the extinction coeffi-
`cient and is dependent on the wavelength of the light passing
`through the substance and on the chemical nature of the
`substance. C is the concentration of the substance and Lis
`the path length of the absorbing material.
`
`[0046] The present invention makes use of just one of the
`wavelengths from the pulse oximeter probe, since the objec-
`tive is to observe only relative changes in the pulse wave
`shape, which in turn is derived from systolic blood volume
`changesin the finger. Since a pulse oximeterprobe is part of
`all portable sleep diagnostic screening devices, it is a further
`object of the present invention to tap into the received light
`intensity signal of an existing probe, thereby alleviating the
`need for any additional patient sensors.
`
`[0047] FIG. 3 shows a typical peripheral pulse waveform.
`Pulse height
`is the number of A/D counts between the
`minimum and maximum excursions of each pulse, while the
`slope is also calculated in A/D counts for a fixed period of
`time beginning about 40 ms after a minimum is detected.
`
`[0048] The first and second derivative waveforms ofthe
`photoplethysmographic waveform have characteristic con-
`tours, and the contour of the second derivative facilitates the
`interpretation of the original waves. The analysis of the
`second derivative ofa fingertip photoplethysmogram wave-
`form has been shown to be a goodindicator ofthe effects of
`vasoconstriction and vasodilation by vasoactive agents, as
`well as an index of left ventricular afterload as shown in
`FIGS. 4A, 4B and 4C.
`
`[0049] FIGS. 4A, 4B and 4C show waveform tracings
`demonstrating the results of administration of vasoactive
`agents. FIG. 4A shows the ECG parameter, FIG. 4B shows
`corresponding PTG and SDPTG waveforms, and FIG. 4C
`showscorresponding AoP and AoFwaveforms. An increase
`in the late systolic component ofaortic pressure (AoP) and
`PTG after intravenous injection of 2.5 mg AGTand a
`deepened d-wave in relation to the height of the a-wave
`(decreased d/s) are seen in SDPTG. Onthe other hand, NTG
`produces marked reduction in late systolic components of
`aortic pressure and PTG, with d-waves becoming shallower
`in relation to the height of a wave (increased d/a). AoF
`indicates ascending aortic flow velocity. Augmentation
`index (AI) is defined as the ratio of the height of the late
`systolic peak to that of the early systolic peak, two compo-
`nents of the ascending aortic pressure at the anacrotic notch.
`
`[0050] Selected Abbreviations and Acronyms
`
`[0051] AGT=Angiotensin
`
`(0052] Al=Augmentation Index
`
`[0055] SDPTG=Second Derivative Wave of Finger-
`tip Photoplethysmography, where the a through d
`components of the second derivative wave are
`described in FIG. 5. The second derivative wave-
`form consists of a, b, c, and d waves in systole and
`an e-wave in diastole.
`
`technology as
`slope (PTS)
`transitional
`Pulse
`[0056]
`applied in the present invention expands on this concept of
`using photoplethysmographically derived waveforms to
`assess changes in vascular tension, whether caused by
`apnaeic obstruction or the more subtle microarousals that are
`not detectable by cortical means. A normalized slope is
`calculated by dividing the height achieved during 40 ms of
`rise time by the maximum height of the pulse waveform
`(=height of late systolic peak). A normalized slope can be
`calculated in real time by a microprocessor controlled device
`as opposedto the post processing (analysis after recording)
`required by secondderivative methods. This will allow use
`of the present
`invention technology in labs performing
`overnight polysomnograph studies in addition to the
`intended use for home sleep screening.
`
`[0057] Since vasoactive drugs have a distinct and prediect-
`able affect on the Al when measured by photoplethysmo-
`graphic methods, by extension the body’s own hormonal
`control of the arterial system shows comparable changes in
`the pulse waveform when measured using similar tech-
`niques.
`
`[0058] The present invention provides a portable, simple,
`and cost effective sleep diagnostic method and apparatus
`capable of detecting arousals and microarousals without
`adding EEG electrodes or additional patient sensors beyond
`those worn during a typical home study.
`
`[0059] Since microarousals have been associated with
`changes in autonomic system outflow, an object of the
`present invention is to provide a small, portable device that
`analyzes the shape of the arterial
`finger pulse,
`thereby
`detecting on a beat by beat basis changes in vascular tone
`directly attributable to microarousals. The present invention
`uses a photoplethysmographically derived arterial blood
`volume waveform for monitoring changein peripheral arte-
`rial vascular tone in conjunction with A/D converters and a
`microcontroller for analyzing the morphology of the pulsa-
`tile signal.
`
`inven-
`[0060] Detection of microarousals by the present
`tion compares favorably with results achieved using pulse
`transit
`time (PTT) devices, EEG analysis, ECG analysis,
`esophagal pressure (Pes), and combinations of these meth-
`ods. Although PTTand peripheral arterial tonometry (PAT)
`have both been receiving much attention as techniques for
`detecting changes in the ANS during sleep studies, PATis
`relatively expensive and PTT has implementation problems
`caused by motionartifact.
`
`[0061] Efficacy of the present invention has been verified
`through monitoring of test subjects performing a “Valsalva
`Maneuver,” which is the quickest and most dramatic method
`of producing ANS discharge—a resulting increase in intra-
`pulmonic pressure produced by forcible exhalation against
`
`012
`
`012
`
`

`

`US 2004/0059236 Al
`
`Mar. 25, 2004
`
`the closed glottis. This produces a sympathetic discharge
`with subsequent vascular constriction.
`
`[0062] A typical response to the Valsalva Maneuver is
`shown in FIG, 6. The normalized slope increases signifi-
`cantly, around 30%on the average which we postulate to be
`caused by increased rate of heart
`tissue conduction,
`increased contraction force, and increased rigidity in the
`arterioles. FIG. 6 shows changes in Normalized Slope
`produced by the present invention during a Valsalva Maneu-
`ver. The increase in ANS outflow begins around heart beat
`59, indicated by the sharp rise in the normalized slope of he
`pulsatile arteriole waveform.
`
`Further testing was conducted using daytime nap
`(0063]
`studies—Several short daytime nap studies were performed
`on sleep deprived volunteers for the purpose of scoring the
`sleep stages during these naps and looking for correlations
`between the stages and recorded normalized PTS slopes.
`None of the subjects were known to have sleep disordered
`breathing. Volunteers were monitored with two central lobe
`electroencephalographic EEGelectrodes, two occipital EEG
`electrodes, two electrooculogram (EOG) electrodes, a chin
`electrode, a nasal air flow device, two respiratory airflow
`belts, and a PTS apparatus of the present invention, which
`provided a normalized slope value on a beat to beat basis.
`
`[0064] A typical recording of the normalized slope (on a
`scale of 0 to 100, where 100 is vertical) versus the sleep
`stages is shown in FIGS. 7A and 7B. The sleep stages were
`scored by a registered polysomnographic technologist
`(RPSGT) from the EEG, EOG, and respiratory waveforms.
`FIGS. 7A and 7B showa sleep stage hypnogram of an hour
`and a quarter sleep study. FIG. 7A shows sleep ratio
`percentages through the duration of the study. FIG. 7B
`showsa graph that has been scored from EEG, EOG, and
`respiratory waveforms according to the sleep scoring con-
`vention of the American Sleep Academy. Point A is the
`beginning of stage 3 sleep, corresponding to point B on the
`normalized pulse slope diagram. Area C is stage 4, and a
`definitive corresponding area of reduced slope values can be
`seen in the area labeled D. As sleep becomeslighter, rising
`from at point E to stage 3 and then stage 2, a corresponding
`rise in slope can be seen starting at point F.
`
`(0065] A block diagram of a preferred embodimentof the
`present invention apparatus is shown in FIG, 8. The device
`is battery powered81, with sufficient capacity for a 1