Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02318735 2000-07-25
WO 99/39627 PCTNS98/02037
METHOD AND APPARATUS FOR
NON-INVASIVE DETERMINATION OF GLUCOSE IN BODY FLUIDS
FIELD OF THE INVENTION
The present invention relates to non-invasive methods and
devices for determining the level of glucose in a body fluid of a subject.
BACKGROUND OF THE INVENTION
There are numerous reasons for determining the level of glucose
present in body fluid of a subject. .In the case of a person suffering from
diabetes, it is often necessary to determine the glucose level in blood daily,
or
10 even more frequently. Non-invasive approaches to determination of blood
glucose levels have been suggested in the patent literature. For example,
United States Patent No. 5,036,861 (issued to Sembrowich ef al, on August 6,
1991 } describes a wrist-mountable device having an electrode which measures
glucose present in sweat at the skin surface. United States Patent No.
5,222,496 (issued to Clarks et al. on June 29, 1993} describes an infrared
glucose sensor mountable, for instance, on a wrist or finger. United States
Patent No. 5,433,197 (issued to Stark on July 18, 1995} describes
determination of blood glucose through illuminating a patient's eye with near-
infrared radiation. United States Patent Nos. 5,115,133, 5,146,091 and
20 5,197,951 (issued to Knudson on May 19, 1992, September 8, 1992 and
January 19, 1993, respectively) describe measuring blood glucose within
blood vessels of a tympanic membrane in a human ear through light
absorption measurements. The specifications of all of these patents are
incorporated herein by reference.
25 The most common current approaches to determining blood
glucose levels still appear to involve obtaining a sample of the person's
blood
and then measuring the level of glucose in the sample. These approaches will
not be reviewed here except to say that obtaining the blood sample
necessarily involves an invasive technique. Generally, the person's skin is
30 broken or lanced to cause an external flow of blood which is collected in
some
fashion for the glucose level determination. This can be both inconvenient and
SU8ST1TUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99/39627 PGTNS98/02037
-2-
distressful for a person and it is an object of the present invention to avoid
the
step of obtaining a blood sample directly, at least on a routine or daily
basis.
It is known that skin tissue, when immersed in an aqueous
glucose solution, equilibrates linearly with the concentration of external
glucose ("Glucose entry into the human epidermis. I. The Concentration of
Glucose in the Human Epidermis", K.M. Halprin, A. Ohkawara and K. Adachi,
J. Invest. Dermatol., 49(6): 559, 1967; "Glucose entry into the human
epidermis. II. The penetration of glucose into the human epidermis in vitro",
K.M. Halprin and A. Ohkawara, J. Invest. Derm., 49(6): 561, 1967). It has also
been shown that skin glucose can vary in synchrony with blood level glucose
during standardized tolerance testing in vivo ("The cutaneous glucose
tolerance test I. A rate constant formula for glucose disappearance from the
skin", R.M. Fusaro, J.A. Johnson and J.V. Pilsum, J. Invest Dermatol., 42:
359, 1964; "The cutaneous glucose tolerance test", R.M. Fusaro and J.A.
Johnson, J. Invest. Dennatol ., 44: 230, 1965). It is also known for
equilibration of glucose levels to occur between blood and interstitial fluids
in
contact with blood vessels ("A microdialysis method allowing characterization
of intercellular water space in human", P. Lonnroth, P.-A. Jansson and U.
Smith, The American Journal of Physiology, 253 (Endocrinol. Metab., 16):
E228-E231, 1987; "Assessment of subcutaneous glucose concentration;
validation of the wick technique as a reference for implanted electrochemical
sensors in normal and diabetic dogs,u U. Fischer, R. Ertle, P. Abel, K.
Rebrin,
E. Brunstein, H. Hahn von Dorsche and E.J. Freyse, Diabetologia, 30: 940,
1987). Implantation of dialysis needles equipped with glucose sensors has
shown that orally ingested glucose load is reflected by parallel changes in
skin
tissue glucose.
Radio frequency spectroscopy using spectral analysis for in vitro
or in vivo environments is disclosed in WO 9739341 (published October 23,
1997) and WO 9504496 (published February 16, 1995). Measurement of a
target chemical such as blood glucose is described.
SUBSTITUTE SHEET (RULE 26~
CA 02318735 2000-07-25
WO 99/39627 PC'f/US98/02037
-3-
SUMMARY OF THE INVENTION
The present invention is a method and apparatus for non-
invasively monitoring levels of glucose in a body fluid of a subject.
Typically,
blood glucose levels are determined in a human subject.
In a preferred embodiment, the invention is a method for non-
invasively monitoring glucose in a body fluid of a subject in which the method
includes steps of measuring impedance between two electrodes in conductive
contact with a skin surface of the subject and determining the amount of
glucose in the body fluid based upon the measured impedance. Typically, the
body fluid in which it is desired to know the level of glucose is blood. In
this
way, the method can be used to assist in determining levels of insulin
administration.
The step of determining the amount of glucose can include
comparing the measured impedance with a predetermined relationship
between impedance and blood glucose level, further details of which are
described below in connection with preferred embodiments.
In a particular embodiment, the step of determining the blood
glucose level of a subject includes ascertaining the sum of a fraction of the
magnitude of the measured impedance and a fraction of the phase of the
measured impedance. The amount of blood glucose, in one embodiment, is
determined according to the equation: Predicted glucose = (0.31 ) Magnitude +
(0.24)Phase where the impedance is measured at 20 kHz.
In certain embodiments, impedance is measured at a plurality of
frequencies, and the method includes determining the ratio of one or more
pairs of measurements and determining the amount of glucose in the body
fluid includes comparing the determined ratios) with corresponding
predetermined ratio(s), i.e., that have been previously correlated with
directly
measured glucose levels.
In certain embodiments, the method of the invention includes
measuring impedance at two frequencies and detemlirtingt~he amount of
glucose further includes determining a predetermined index, the index
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99/39627 PCTNS98J02037
-4-
including a ratio of first and second numbers obtained from first and second
of
the impedance measurements. The first and second numbers can include a
component of said first and second impedance measurements, respectively.
The first number can be the real part of the complex electrical impedance at
the first frequency and the second number can be the magnitude of the
complex electrical impedance at the second frequency. The first number can
be the imaginary part of the complex electrical impedance at the first
frequency
and the second number can be the magnitude of the complex electrical
impedance at the second frequency. The first number can be the magnitude of
the complex electrical impedance at the first frequency and the second number
can be the magnitude of the complex electrical impedance at the second
frequency. In another embodiment, determining the amount of glucose further
includes determining a predetermined index in which the index includes a
difference between first and second numbers obtained from first and second of
said impedance measurements. The first number can be the phase angle of
the complex electrical impedance at the first frequency and said second
number can be the phase angle of the complex electrical impedance at the
second frequency.
The skin site can be located on the volar forearm, down to the
wrist, or it can be behind an ear of a human subject. Typically, the skin
surtace is treated with a saline solution prior to the measuring step. An
electrically conductive gel can be applied to the skin to enhance the
conductive contact of the electrodes with the skin surface during the
measuring step.
The electrodes can be in operative connection with a computer
chip programmed to determine the amount of glucose in the body fluid based
upon the measured impedance. There can be an indicator operatively
connected to the computer chip for indication of the determined amount of
glucose to the subject. The indicator can provide a visual display to the
subject.
SU8ST1TUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99139b27 PCT/US98/0203?
-5-
In certain embodiments, the computer chip is operatively
connected to an insulin pump and the computer chip is programmed to adjust
the amount of insulin flaw via the pump to the subject in response to the
determined amount of glucose.
Electrodes of a probe of the invention can be spaced between
about 0.2 mm and about 2 cm from each other.
In another aspect, the invention is an apparatus for non-invasive
monitoring of glucose in a body fluid of a subject. The apparatus includes
means for measuring impedance of skin tissue in response to a voltage
applied thereto and a microprocessor operatively connected to the means for
measuring impedance, for determining the amount of glucose in the body fluid
based upon the impedance measurement(s). The means for measuring
impedance of skin tissue can include a pair of spaced apart electrodes for
electrically conductive contact with a skin surtace. The microprocessor can be
programmed to compare the measured impedance with a predetermined
correlation between impedance and blood glucose level. The apparatus can
include means for measuring impedance at a plurality frequencies of the
applied voltage and the programme can inGude means for determining the
ratio of one or more pairs of the impedance measurements and means for
comparing the determined ratios) with corresponding predetermined ratios) to
determine the amount of glucose in the body fluid.
The apparatus preferably includes an indicator operatively
connected to the microprocessor for indication of the determined amount of
glucose. The indicator can provide a visual display for the subject to read
the
determined amount of glucose. It is possible that the indicator would indicate
if
the glucose level is outside of an acceptable range.
In a particular embodiment, the microprocessor is operatively
connected to an insulin pump and the apparatus includes means to adjust the
amount of insulin flow via the pump to the subject in response to the
determined amount of glucose.
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99/39627 PCTNS98/02037
-6-
The apparatus can include a case having means for mounting the
apparatus on the forearm of a human subject with the electrodes in
electrically
conductive contact with a skin surtace of the subject.
In a particular embodiment, the apparatus includes means for
calibrating the apparatus against a directly measured glucose level of a said
subject. The apparatus can thus include means for inputting the value of the
directly measured glucose level in conjunction with impedance measured
about the same time, for use by the programme to determine the blood glucose
level of that subject at a later time based solely on subsequent impedance
measurements.
A microprocessor of the apparatus can be programmed to
determine the glucose level of a subject based on the sum of a fraction of the
magnitude of the measured impedance and a fraction of the phase of the
measured impedance. In a particular embodiment, the apparatus is set to
measure impedance at 20 kHz and the microprocessor is programmed to
determine the glucose level of a subject based on the equation: Predicted
glucose = (0.31 )Magnitude + (0.24) Phase.
BRIEF DESCRIPT10N OF THE DRAWINGS
Preferred embodiments of the invention will now be described,
reference being had to the accompanying drawings, wherein:
Figure 1 shows plots of various indices as a function of time and
glucose concentration based on impedance measurements taken on the skin
(SCIM) of a first diabetic subject. Figure 1 (a) shows MIX versus measurement
number, the timing of the measurements being given in Table 1. Figure 1 (b)
shows PIX versus measurement number. Figure 1 {c) shows RIX versus
measurement number. Figure 1 (d) shows IMIX versus measurement number.
The determinations of MIX, PIX, RIX and IMIX are described in the text.
Figures 2(a), 2(b), 2(c) and 2(d) are similar to Figures 1 (a) to
1 (d), respectively, but are based on impedance measurement taken on the
skin of a second diabetic subject.
SU6ST1TUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99/39627 PCT/US98/02037
-7-
Figure 3 is a plot showing the reading (average of ten readings)
of a dermal phase meter as a function of directly determined blood glucose
concentration. Measurements were taken on a site on the left forearm (~) and
right forearm (+); and
Figure 4 is similar to Figure 3, but readings were taken at a
finger.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
A preferred method of the invention involves directly contacting a
subject's skin with an electrode, taking one or more impedance measurements
and determining the subject's blood glucose level based on the impedance
measurement(s). Preferably, there is a computer programmed to make the
determination based on the impedance measurement(s). In one aspect, the
invention includes deriving a number of indices from one or more
measurements of impedence between poles of the electrode. The values) of
the one or more indices is an indicator of, i.e. correlates with, the
subject's
blood glucose level.
Thus, the invention is illustrated below by laboratory feasibility
tests to establish that a correlation between one or more such index values
based on impedance measurements) and a subject's blood glucose level
exists. The tests were conducted using particular parameters, for example
impedance measurements obtained at a certain frequency or certain
frequencies, and particular indices were dervied therefrom. It will be
understood that other andlor additional frequencies may be found to be more
optimal and that other indicies may well be found to be more optimal.
Examples
Each of two subjects was treated as indicated in Table 1.
Impedance measurements were taken at the volar forearm using the "SCIM"
apparatus described below. Impedance measurements were taken at thirty-
one frequencies and four different indices were determined using two of the
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 9913962? PCT/US98/02037
_$_
frequencies: 20 and 500 kHz. Directly measured blood glucose levels of each
subject are indicated in Table 1.
Table 1:
Treatment
Regimen
of Subjects
Measurement Blood Glucose Blood Glucose
No. -
time (minutes) Measurement Measurement
First Subject Second Subject
0 0 154 141
Ingest 50 g glucose
1 10 146 164
2 20 174 194
3 30 246 232
4 40 228 257
Ingest 50 g glucose
3
5 50 268 304
6 60 255 348
7 70 320 346
8 80 320 355
9 90 399 361
10 100 343 383
11 110 334 381
Rapid insulin 4 units 8 units
administered
12 125 358 379
13 140 377 346
14 155 353 333
Four indices, MIX, PIX, RIX and IMIX were determined (see
below) and plotted as a function of time. Results are shown in Figures 1 and
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99/39627 PCT/US98/02037
-9-
2, the data collected prior to the first glucose ingestion being assigned "0"
on
E the x-axis of each plot.
Spearman rank order correlation coefficients were determined,
and are presented Table 2 and 3 for the first and second subjects,
respectively. A value of Ps0.05 is often considered to be a satisfactory
correlation. As can be seen in Table 2, a satisfactory correlation was
obtained
for both the MIX and the 1MIX indices for the first subject. As can be seen in
Table 3, a satisfactory correlation was obtained for the MIX, PIX and IMIX
indices for the second subject. The value of P for the RIX index was very
close to being satisfactory. It must be borne in mind that these values were
obtained from a small sample set and yet a clear indication of a satisfactory
correlation for more than one index has been obtained in these experiments.
Optimization of the parameters of frequency and the choice of index or indices
might well lead to a significant improvement on the results given here.
Table 2: Statistical
Analysis of Relationship
between Measured
Glucose
Levels and Selected
Indices for f=irst
Subject
Spearman
Rank
Order
Correlations
Pair of Variables Valid Spearman t(N-2) P
N R
Glucose Level & 15 -.722719 -3.77028 0.002336
MIX
Glucose Level 8~ 15 .865832 6.23942 0.000030
PIX
Glucose Level & 15 -.418980 -1.66372 0.120073
RIX
Glucose Level & 15 -.710833 -3.64385 0.002972
IMIX
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99l396Z7 PGT/US98/OZ037
-10-
Table 3: Statistical
Analysis of Relationship
between Measured
Glucose Levels
and Selected Indices
for Second Subject
Spearman
Rank
Order
Conrelations
Pair of Variables Valid Spearman t(N-2) P
N R
Glucose Level & 15 -.616622 -2.82405 0.014353
MIX
Glucose Level & 15 .266547 .99712 0.336903
PIX
Glucose Level & 15 -.477094 -1.95731 0.072133
RIX
Glucose Level & 15 -.607686 -2.75888 0.016260
IMIX
The impedance measurements on which the results shown in
Figures 1 and 2 are based were obtained using a Surtace Characterizing
Impedance Monitor (SCIM) developed by Ollmar (United States Patent No.
5,353,802, issued October 11, 1994; "Instrument evaluation of skin
irritation",
P.Y. Rizvi, B.M. Morrison, Jr., M.J. Grove and G.L. Grove, Cosmetics &
Toiletries., 111: 39, 1996; "Electrical impedance index in human skin:
Measurements after occlusion, in 5 anatomical regions and in mild irritant
contact dermatitis", L. Emtestam and S. Ollmar, Conf. Derm. 28: 337, 1975;
"Electrical impedance for estimation of irritation in oral mucosa and skinu,
S.
Ollmar, E. Eek, F. Sundstrorn and L. Emtestam, Medical Progress Through
Technology, 21: 29, 1995; "Electrical impedance compared with other non-
invasive bioengineering techniques and visual scoring for detection of
irritation
in human skin", S. Ollmar, M. Nyren, I: Nicander and L. Emtestam, Brit. J.
DermafoG 130: 29, 1994; "Correlation of impedance response patterns to
histological findings in irritant skin reactions induced by various
surtactants", 1.
Nicander, S. Ollmar, A. Eek, B. Lundh Rozell and L. Emtestam, Brit. J.
Dermatol. 134: 221, 1996) which measures bioelectrical impedance of the skin
at multiple frequencies. The instrument is basically an AC-bridge fabricated
from standard laboratory instruments: a function generator, a digital
oscilloscope, impedance references, and a driver for the probe.
The indices plotted in Figures 1 and 2 were determined as
follows:
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99/39627 PCT/US98/02037
-11 -
MIX (magnitude index) = abs(Z2~)labs(Z)
PIX (phase index) = arg{Z~) - ar~g(Z~,~)
RIX (real part index) = Re(Zz~~abs(Z)
IMIX (imaginary part index) _ !m(Z~~abs(Z)
where abs(Z,) is the magnitude {modulus) of the complex electrical impedance
at the frequency i, arg(Z,) the argument (phase angle) in degrees, Re(Z,) the
real part of the complex electrical impedance, and Im(Z,) the imaginary part
of
the complex electrical impedance. The magnitudes and phase angles are
delivered by the instrument, and the real and imaginary parts are calculated
according to the elementary complex number relationships: Re(Z,) _
abs(Z,)'cos[arg(Z,)] and Im(Z,) = abs(Z,)'sin[arg(Z;)~.
The RIX reflects changes mainly in conductivity; the IMIX reflects
mainly reactance changes, which are of capacitfve nature; the MIX reflects
changes along the length of the vector describing the impedance in complex
space, which will be emphasized if the real and imaginary parts change in the
same direction and proportion; the PIX will be emphasized if the real and
imaginary parts change in different directions and/or in different
proportions.
Prior to contacting a subject's skin with the electrode, the skin is
treated with a 0.9°~ saline solution by holding a soaked gauze against
the
measurement site for about a minute and then wiping the site with a dry cloth.
The purpose of this step is to ensure adequate electrical coupling between the
skin and the probe (electrode) in order to reduce variability that may
introduced into the measurements by stratum corneum. A person skilled in the
art would understand that variations are possible, and more optimal pre-
treatment conditions may be obtainable.
Blood glucose levels were determined directly from a blood
sample using a faucet pr7ck and measuring the blood glucose concentration
with an Elite Glucometer according to manufacturer's instructions {Elite
Glucometer, Miles Canada, Diagnostics Division, Division of Bayer).
30 In a second set of experiments, 31 subjects were tested using the
SCIM apparatus. A baseline measurement was taken and standardized food
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99139627 PCT/US98102037
-12-
packet ingested. Two additional impedance measurements were taken one
half hour and one hour after the initial measurement and blood glucose levels
determined directly. Multiple regression analysis was carried out on data
obtained at 20 kHz and relationship (1 ) established:
Predicted glucose = (0.31 ) Magnitude + (0.24) Phase; F-5.5, p<0.005
The multiple R for the prediction was 0.33.
The SCIM instrument was used to measure impedance measured
at 31 different frequencies logarithmically distributed in the range of 1 kHz
to 1
Mhz (10 frequencies per decade). Subsequent determinations were based, in
the first set of experiments, on two of the frequencies: 20 and 500kHz; and in
the second set of experiments, 20 kHz only. It may be found in the future that
there is a more optimal frequency or frequencies. It is quite possible, in a
commercially acceptable instrument that impedance will be determined at at
least two frequencies, rather than only one. For practical reasons of
instrumentation, the upper frequency at which impedance is measured is likely
to be about 500 kHz, but higher frequencies, even has high as 5 MHz or higher
are possible and are considered to be within the scope of this invention.
Relationships may be established using data obtained at one, two or more
frequencies.
It may be found to be preferable to use an artificial neural
network to perform a non-linear regression.
A preferred instrument, specifically for determining glucose levels
of a subject, includes a 2-pole measurement configuration that measures
impedance at multiple frequencies, preferably two well spaced apart
frequencies. The instrument includes a computer which also calculates the
index or indices that correlate with blood glucose levels and determines the
glucose levels based on the corrlelation(s).
The invention is also illustrated by experimer>~ that were carried
out with a dermal phase meter (DPM) available from Novas Technology
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99/39617 PCT/US98/02037
-13-
Corporation of Gloucester, Massachusetts. Measurements were taken with the
dermal phase meter at two skin sites, the forearm and the middle finger. The
scale of the meter is from 90 to 999. It is thought that a higher reading
indicates a higher degree of skin hydration. Blood glucose measurements
were also measured directly (MgsIdL) using an Elite Glucometer determined
directly from a blood sample using a lancet prick and measuring the blood
glucose concentration according to manufacturer's instructions (Elite
Glucometer, Miles Canada, Diagnostics Division, Division of Bayer). Typical
results are shown in Figures 3 and 4. Measurements were taken at various
times to track changes in skin hydration from that present while fasting
overnight, attending ingestion of a typical meal for breakfast or lunch and
following a peak of blood glucose and decline to about 100 Mgs/dL.
In these experiments, a probe sensor was placed against the skin
surface and held lightly until the instrument indicated completion of data
acquisition. Time interval (latch time} for data acquisition was selected at
zero
seconds (instantaneous). Other suitable time periods can be anywhere 0 and
30 seconds, or between 0.5 and about 10 seconds, or between about 1 and 5
seconds or about 5 seconds. The results obtained using the dermal phase
meter are plotted as function of blood glucose concentration in Figures 3 and
4, respectively. Each plotted point represents the average of 10
measurements using the dermal phase meter. Studies were pertormed in the
morning on fasting subjects. After baseline measurements on fasting, food
was ingested to raise blood glucose levels. Studies continued until blood
glucose levels declined to baseline levels.
Figures 3 and 4 indicate that the NovaT'~' meter reading of the
skin increases with increasing blood glucose concentration.
In one aspect of the invention, electrodes of a device are placed
in conductive contact with a subject's skin in order to measure impedance (Z)
at various frequencies (f) in a range from a few Hertz (hz) (say 10 hz) to
about
5 Mhz. A more typical range would be between 1 kHz and 1 Mhz, and more
likely between 5 kHz and 500 kHz. Electrodes of the device are electrically
SUBSTITUTE SHEET (RULE 26)
CA 02318735 2000-07-25
WO 99/39627 PC'fNS98102037
-14-
connected to a metering device which indicates the impedance at a selected
frequency of applied voltage, as understood by a person skilled in the art. In
a
particular embodiment, the device is programmed to operate at the selected
frequencies in rapid sequence. Alternative modes of operation are possible,
for example, the voltage can be rapidly increased with time and Fourier
transformation carried out to obtain a frequency spectrum. Ratios of
impedance measured at various frequencies are determined and the blood
glucose level of the subject is measured directly. This process is repeated at
different times so as to make the determination at a number of different
glucose levels. In this way, ratios of impedance determined at particular
frequencies which are found to reproducibly reflect a person's blood glucose
levels over a range of glucose levels are determined. The ratios of measured
impedance at the selected frequencies can thus be correlated with directly
measured glucose levels, that is, a plot in which log(Z~IZZ) vs log (f) is a
linear
correlation, or an approximately linear correlation, is determined. This
relationship is then used to determine the blood glucose level of the person
directly from ratios of similarly obtained impedance measurements, thus
avoiding an invasive technique for obtaining the blood glucose level.
Impedance includes both resistance and reactance.
It may be found for a proportion of the population that there is a
universal set of impedance frequency ratios, thus avoiding the necessity of
determining individual correlations.
The general approach described for the foregoing aspect of the
invention can be used in connection with other indices based on impedance
measurements, such as MIX, PIX, RIX and IMIX described above.
It is important, of course, to be able to reliably reproduce results
as much as possible in order for this type of device to be useful. To this end
an appropriate skin site is chosen. Generally speaking, an undamaged skin
site and one that is not heavily scan-ed would be chosen. A skin site having a
stratum corneum which is less likely to deleteriously intertere with the
measurements is chosen. A likely possibility is the volar forearm, down to the
SUBSTITUTE SHEET (RULE 28)
CA 02318735 2000-07-25
WO 99/39627 PCTIUS98102037
_15_
wrist, or behind an ear. The skin surtace can be treated just prior to
measurement in order to render the stratum corneum more electrically
transparent by application, for example, of a physiological saline dressing
for
about a minute. Excess liquid should be removed before application of the
probe.
Given the importance of reliable glucose level determinations in
setting insulin administrations, it is important that the invention be used
only in
circumstances in which it is known that the approach described herein reliably
indicates glucose levels of a subject. It is possible that the invention would
not
be suitable for use with a given proportion of the population or 100% of the
time with a given individual. For example, an individual may have a skin
condition which deleteriously interferes with impedance measurements,
making it difficult to assume that impedance measurements can reliably
indicate a person's blood glucose level. For such a person, a different
approach to glucose level determination would be more suitable.
An apparatus that utilizes a neural network to carry out analyses
based on impedance could be trained for a specific subject, or possibly a
group of subjects. An example of such a group of subjects might be subjects of
the same sex, belonging to a particular age group and within particular height
and weight ranges.
~ may be advantageous to optimize the spacing of the electrodes
of the probe. That is, it may found that the electrodes of a SCIM probe are
too
close to each other to provide maximally reproducible results. An object of a
suitable probe is to have electrodes spaced from each other to obtain optimal
penetration of current into tissue containing glucose in its interstitial
spaces. It
is expected that electrodes spaced somewhere between about 0.2 mm and
about 2 cm are suitable.
Additionally, the use of a gel can improve skin :probe contact to
more reliably produce useful measurements, as would be known to a person
skilled in the art, e.g., a gel comprising mostly water in combination with a
SUBSTITUTE SHEET (RULE 2b)
CA 02318735 2000-07-25
WO 99/39627 PCT/US98l02037
-16-
thickener such as Cellusize, glycerin or propylene glycol as a moisturizer,
and
a suitable preservative.
An apparatus for non-invasive monitoring of glucose in a body
fluid of a subject includes means for measuring impedance of skin tissue in
response to a voltage applied thereto, i.e. a probe. There is a computer
processor operatively connected to the means for measuring impedance for
determining the blood glucose level based upon one or more impedance
measurements. The microprocessor is programmed to calculate the blood
glucose level of a subject based upon impedance measurements taken at one
or more frequencies. In a particular embodiment, a calcuation based upon
impedance at a single frequency, along the lines of that shown in relationship
(1 ), is carried out by the processor. In another embodiment, the calculation
includes determining MIX and/or IMIX. The calculation might include
determining PIX. The calculation might include determining R1X. It might be
necessary to calibrate an individual apparatus for use with a particular
subject.
In such case, the apparatus includes means for calibrating the apparatus
against a directly measured glucose level of that subject. The apparatus could
thus include means for inputting the value of the directly measured glucose
level in conjunction with impedance measured about the same time, for use by
the programme to determine the blood glucose level of that subject at a later
time based solely on subsequent impedance measurements.
In one embodiment, a meter is worn in which a probe is
continuously in contact with the skin and moisture buildup between occlusive
electrodes and the skin is sufficient to obtain useful measurements. The
device can be mountable on a person's forearm, much like a wristwatch. Such
an embodiment might not prove to be useful for all subjects.
As previously stated, it might be found to be necessary for a
meter to be calibrated individually, that is, it might be necessary to
determine
the relationship between ascertained impedance ratios or index or indices of
interest, and blood glucose levels of an individual and base the operation of
the particular meter for that individual on the relationship. To this end, a
SUBSTnUTE SHEET (RULE 26~
CA 02318735 2000-07-25
WO 99/39627 PCTNS98/02037
-17-
preferred monitoring device of the invention includes means for calibrating
the
relationship between a directly measured blood glucose level and an index or
indices of interest.
Because blood glucose level determinations of the present
invention are non-invasive and relatively painless it is possible to make such
determinations with a greater frequency than with a conventional pin-prick
method. In a particularly advantageous embodiment, blood glucose levels are
monitored quite frequently, say every fifteen or five, or even one minute or
less, and an insulin pump is interfaced with the meter to provide continual
control of blood glucose in response to variations of blood glucose levels
ascertained by means of the meter.
The disclosures of all references, and particularly the
specifications of all patent documents, referred to herein, are incorporated
herein by reference.
The invention now having been described, including the best
mode currently~known to the inventors, the claims which define the scope of
the protection sought for the invention follow.
SU8ST1TUTE SHEET (RULE 26)