48

Current Diabetes Reviews, 2012, 8, 48-54

Non-Invasive Methods of Glucose Measurement: Current Status and Future Perspectives Andreea Ciudin, Cristina Hernández and Rafael Simó* CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Instituto de Salud Carlos III (ISCIII), Diabetes and Metabolism Research Unit, Vall d’Hebron Institut de Recerca (VHIR), Hospital Universitari Vall d’Hebrón, Universitat Autònoma de Barcelona, Barcelona, Spain Abstract: Diabetes mellitus (DM) is a very common disease which, if a good glycemic control is not achieved, can lead to serious chronic complications such as cardiovascular disease, retinopathy, nephropathy and neuropathy. Selfmonitoring of blood glucose is a fundamental tool for the proper adjustment of the treatment of diabetes and, at present, it is based mainly on capillary blood obtained by finger-prick (the classical glucometers). Since this method is painful and the strips are expensive, investigators have been attracted by the idea of using a non-invasive device for determining blood glucose which would permit more frequent testing and a tighter control of diabetes. The non-invasive measurement of blood glucose is based on the ability of the glucose molecule to interact with various chemical or physical methods. Nevetheless, in spite of some encouraging results and the efforts made over the past 30-40 years, there is no device available at present for use in clinical practice. A possible explanation might be the combination between the specific features of each method and the specific characteristics of diabetic patients, which make them respond differently to physical and chemical methods when compared to their healthy counterparts. In this paper we will give an overview of the noninvasive devices tested so far and their implications for the clinical management of the diabetic patient.

Keywords: Blood glucose self-monitoring, diabetes mellitus, non-invasive glucose measurement. 1. INTRODUCTION Diabetes mellitus (DM) is a very common disease that can lead to serious vascular complications (cardiovascular disease, retinopathy, nephropathy and neuropathy). Good glycemic control is important since it has been shown to prevent or arrest the progression of the chronic complications of diabetes [1-3]. Self-monitoring of capillary blood glucose by the patient is a fundamental tool for the proper adjustment of the treatment of diabetes, especially insulin. The recommendations of the American Diabetes Association [4] include at least 3 self-determinations of capillary blood glucose for diabetic patients using multiple insulin injections or insulin pump therapy. It also suggests that self-monitoring blood glucose may be useful in order to achieve the metabolic control targets in the rest of the diabetic patients, a situation which represents an important economic burden. The devices currently used are mainly based on capillary blood obtained by finger-prick (the classical glucometers). In addition, in recent years methods based on subcutaneus interstitial fluid glucose measurements (minimal invasive methods) have emerged. Currently, the gold standard for measuring blood glucose is the glucose-oxidase method. Obviously, the finger-pricking is painful and the strips are expensive, and replacing it would represent a substantial improvement in the quality of life of diabetic patients and a possible reduction in the economical burden. Investigators have been attracted by the idea of a non-invasive (NI) *Address correspondence to this author at the Diabetes and Metabolism Research Unit, Institut de Recerca Hospital Universitari Vall d’Hebron, Pg. Vall d’Hebron 119-129, 08035 Barcelona, Spain; Tel: 34 934894172; Fax: 34 934894032; E-mail: [email protected] 17-/12 $58.00+.00

method to determine blood glucose which would permit more frequent testing and a tighter control of diabetes. The first device proving that the NI determination of glucose in vivo was possible was reported in 1982 by Rabinovitch et col. In this experiment, blood glucose levels were estimated by measuring the optical rotation of the aqueous humor with scleral lens [5] and since then many attempts have been made to find a less cumbersome method. Nevertheless, currently there is no device approved for clinical practice. In this paper we will briefly summarize the methods tested so far and their implications in the clinical management of the diabetic patient. 2. PRINCIPALS MEASURING

OF NON-INVASIVE GLUCOSE

Non-invasive determination of blood glucose is a very challenging but difficult task, since an ideal device should meet several conditions, such as being totally non-invasive, without producing any kind of lesion in the skin or other body barrier and be able to properly detect glucose concentrations even in situations of rapid blood glucose changes [6,7]. Glucose can be found in several compartments and body fluids, besides blood, such as interstitial fluid, tears, vitreous fluid, urine and sweat, and many methods are based on measuring the glucose in these compartments. Ideally, the technique should detect glucose concentration at any time and in any condition, showing identical and simultaneous variations in glucose concentration when referred to blood content. From the studies so far it has been demonstrated that there is a specific lag in the equilibration between blood © 2012 Bentham Science Publishers

Non-Invasive Methods of Glucose Measurement

glucose values and glucose concentrations in other body compartments [8-11]. Basically, the NI methods used to determine the blood glucose levels can be classified into two categories according they track intrinsic or extrinsic properties of glucose molecule. The glucose molecule can interact with various chemical or physical methods independently of the body compartment (the intrinsic property) or it can induce tissue specific local changes (the extrinsic property). In order to facilitate a better understanding, in this review we will use this classification. Nevertheless, there are authors who prefer to classify the NI glucose determinations into optical and non-optical, since the optical properties of glucose are rather specific and these methods have shown better results and a better correlation with blood glucose content [11]. In order to facilitate a better understanding we will review the methods using the first criteria. 2.1. Methods Based on Extrinsic Properties of Glucose As explained above, these methods are tissue compartment dependent, since they are based on tracking the changes produced at local level by glucose concentration and variations. 2.1.1. Light Scattering Coefficient Some methods are based on estimating blood glucose by detecting the light scattering coefficient of a tissue compartment, known as optical coherence tomography and temperature modulated localized reflectance. The theoretical base is that in the case of a heterogeneous media such as, for example, human skin, the scattering of light depends of the balance between the refractive index of the medium and the molecules. Basically, changes in the refractive index between the particles (for instance glucose) and the interstitial fluid lead to an alteration in tissue transparency due to an alteration of the scattering coefficient of the tissue. In other words, the scattering coefficient of the skin is influenced indirectly by the refractive indices, and is not a specific feature of the glucose molecule. A number of physiological conditions can also affect it, such as changes in body temperature (as for instance during an inflammatory disease) [12] and shifts in the water/plasma distribution between the intravascular and the interstitial compartments. Due to these limitations, the devices tested so far using this method encounter a serious problem when the stability of the environment is no longer maintained. Signal stability is fundamental in order to realize a correct estimation [12,13]. It should also be taken into account that these methods investigate glucose concentration in the interstitial fluid of the upper dermis of the skin, not directly in the blood. A lag time of 5-30 minutes between interstitial fluid and blood glucose has been demonstrated in several studies [8], and this delay is important especially in the case of sudden changes in blood glucose values and even more when these variations are very large in magnitude [6,7]. 2.1.2. Ultrasound Technology The most widely used technique is photoacoustic spectroscopy, based on ultrasonic waves created by tissue

Current Diabetes Reviews, 2012, Vol. 8, No. 1

49

absorption of pulsating light- laser [14]. The fluid is excited by a short laser pulse with a wavelength that is absorbed by a particular molecule existing in the fluid [15]. Light absorption causes microscopic localized heating in the medium which generates an ultrasound pressure wave which can be detected. This technique limits the possible interference of water, since water has a relatively poor photoacoustic response [16]. On the other hand, one of the limitations could be that when laser light transverses a dense media, like human body compartments, its contribution to the photoacoustic signal is due not only to the absorption coefficient but also the scattering coefficient, resulting in a possible confounding factor. The usual measurement sites are the sclera, fingers or forearm. “Aprise” was a well known non-invasive glucose monitoring device which applies this technique. Apparently the interferences of the confounding factors were avoided by separating the changes in the blood from those in the surrounding tissue compartments, achieving in this way a reliable correlation between the photoacoustic signal that originates in the blood vessel and changes in glucose concentrations [17]. Even if the measured values were assigned to the A and B zones of the Clarke Error Grid, with a global variation of 19% when compared to plasma glucose, the device showed important errors in the case of hypoglycemia and rapid changes in blood glucose levels. 2.1.3. Impedance Spectroscopy Also known as dielectric spectroscopy, this method is based on the hypothesis that changes in blood glucose concentration will induce a decrease in sodium concentration and an increase in potassium levels in red blood cells, thus generating a membrane potential which can be estimated by determining the permittivity and conductivity of the cell membrane through the dielectric spectrum [18,19]. There are several reports describing the use of impedance spectroscopy for non-invasive glucose monitoring [20-23]. Perhaps the best known device based on this tecnique is Pendra® (Pendragon Medical) who was approved for clinical use in May 2003. It was available on the european market for less than two years and subsequently withdrawn for having poor accuracy especially in the hypoglycemia range [24]. Additionally, Caduff et al. [25] reported in 2003 the results of comparing the blood glucose estimated by impedance spectroscopy with both venous blood glucose and capillary glucose determined with a classical glucometer. The results were promising, with 93% of the values situated in the high performance zones (56% and 37% in the A and B zones respectively), but the study was restricted to adult patients with well controlled diabetes without severe diabetic complications. Obviously, the measurements of this technique based on local impedance characteristics, could have been influenced by local changes induced by chronic diabetic complications. Also, the variability of the results due to temperature and the degree of hydration were limiting factors [25]. As seen in the case of the light scattering coefficient, environmental and body composition stability is fundamental in order to realize a correct estimation of blood glucose levels.

50 Current Diabetes Reviews, 2012, Vol. 8, No. 1

Ciudin et al.

2.1.4. Electromagnetic Sensing This is a technique based on the electromagnetic coupling of two inductors for detecting the dielectric parameters of the blood. Glucose can influence the dielectric properties of a given solution, so that in theory an estimation of the glucose concentration can be calculated by this method [22,26]. It seems that, variations in glucose concentrations have a specific voltage signal frequency where the sensitivity is optimal (2, 66 MHz). From the experiments done so far, it is apparent that the variations in the ambient or solution temperature have a strong effect on the dielectric properties. On the other hand, in the case of the blood there are other solutes that could have an influence [27]. To our knowledge there have only been in vitro experiments and it has not yet been tested in humans. 2.2. Methods Based on an Intrinsic Property of Glucose As explained above, there are methods of non-invasive glucose measurement which exploit an intrinsic property of glucose- its ability to interact directly with several physical or chemical principles. Some authors include in this category the fluid harvesting method, which is based on creating micropores on the human skin using a laser light or ultrasound and extracting the glucose particles from the interstitial fluid so making a direct measurement [28]. Reverse iontophoresis is a similar method permitting a direct interaction between glucose molecules and a sensor, by extracting the glucose through the skin using an iontophoretic current [29,30]. Perhaps the best known device so far is the Glucowatch based on reverse iontophoresis, approved by the FDA in 2001 and subsequently withdrawn in 2005 [31] . Since these devices realize a direct measurement of glucose, by extracting the glucose from the body fluids, they can be considered as minimally invasive and so will not be detailed here. Inside this category we can divide the various methods into non-optical and optical in order facilitate understanding, since the optical devices share some common features. 2.2.1. Non-optical Non-invasive Methods Tracking: An Intrinsic Glucose Property

Based

on

2.2.1.1. Ocular Spectroscopy This method is based on the ability of glucose to create reversible covalent bounds with boronic acid [32]. Recent studies have shown that glucose is able to specifically interact with boronic acid even in very small concentration ranges, inferior to 5mM [33]. Lane et al. carried out a study on tear glucose before and after administration of a carbohydrate load, demonstrating that there is a correlation between tear glucose and blood glucose concentration both in normal and diabetic subjects [34]. The blood glucose levels are tenfold higher than the tear glucose levels as shown in some studies [35,36]. Basically, this method is based on using contact lenses with a hydrogel wafer of boronic acid derivatives, interacting reversibly with the glucose molecules present in tears. When the lens is illuminated by a laser light, the reflected light changes its color depending on the intensity of the binding phenomenon, related to tear glucose concentration. Theoretically, a patient

wearing the lenses could compare the color to a precalibrated color strip, by looking into a mirror [33]. The main limitation of this study is the lag time between the tear and blood glucose, of about 10-20 minutes [33, 37]. 2.2.2. Optical Non-Invasive Methods Based on Tracking an Intrinsic Glucose Property 2.2.2.1. Thermal Spectroscopy This method is based on alterations to the thermally modulated optical signals through the physiological and physical effects of blood glucose concentrations. In other words, it is considered that the absorptive effect of glucose emits infrared radiation [11]. Several studies shown that variations in tissue temperature could alter cutaneous vascular and refractive index responses [38], and due to this fact, the method has an important limitation. 2.2.2.2. Fluorescence Spectroscopy This is based on the detection of fluorescence when exciting a glucose solution by an ultraviolet laser light at 308nm. The effect depends on the glucose concentration in the solution [11]. In humans, the fluorescence phenomenon depends not only on glucose concentration, but on skin pigmentations and epidermal thickness [39] and so far its liability has been demonstrated only in vitro experiments. 2.2.2.3. Raman Spectroscopy Raman spectroscopy is a method based on the emission of scattered light produced by the rotations of the molecules existent in a given solution, when excited by a laser light. The degree of vibration and rotation of a molecule in a solution depends on its concentration. Most of the studies on Raman spectroscopy were performed in vitro, using a wave length of the laser light between 200-1800cm-1 and it was seen that Raman bands at 900-1200cm-1 are specific for glucose [40,41]. In vivo studies were performed mainly in animal models, and the eye was the most common site [42], showing a good correlation with glucose concentration. The experiments on human skin showed different results because of the complexity of the lipid structure, which is the main confounding compounds [43]. Nevertheless a recent study [44] on humans, found that the accuracy of the current measurement is approaching the level of clinical usefulness, with 92% of the values situated in the A and B zones of the Clarke Error Grid. But still, one of the main limitations is the size of the device, which makes it impractical to be used in the clinical practice. Also, better calibration and mathematical models are needed in order to eliminate the effect of the other confounding compounds. 2.2.2.4. Mid-infrared Spectroscopy (MIR) This method is based on the ability of glucose molecule to specifically absorb light in the 2500-10000nm spectrum [11]. One advantage of the MIR method is that the signals produced by glucose are sharp and have high absorption coefficient. However, a light penetration of only few micrometers is a significant limiting factor. To overcome this problem measurement by attenuated total reflection has been used. This method uses a beam of light guided through a crystal placed in contact with the skin, thus ensuring that the electromagnetic field created by the reflected light

Non-Invasive Methods of Glucose Measurement

reaches the dermis (the skin layer which contains most glucose). The MIR spectrum is significantly dependent on the water content, making this method very sensitive to the degree of hydration [44,45]. 2.2.2.5. Near-infrared Spectroscopy (NIR) In recent years a special emphasis has been placed on systems of non-invasive measurement of glucose levels in blood based on spectroscopy in the band near infrared (NIR) [46, 47]. This concept is based on the transmission of a band of NIR light through a vascular area of the body (finger, ear, tongue, etc.), and the concentration of glucose in vivo is calculated from the spectral information obtained at the reception. The NIR spectral region has several windows where hemoglobin, melanin and water absorbtion band intensities are low enough to allow light to penetrate into the tissue, enabling NI spectral measurements to be taken. The basic premise of these systems is that specific information about glucose levels is represented in the NIR spectrum of the received signal [48, 49]. In fact, you can obtain glucose measurements in complex biological matrices such as synthetic biological mixtures by the combination of NIR spectroscopy techniques with calibration techniques based on multivariate analysis [50-53]. Despite the great appeal of the application of spectral analysis in the NIR band to obtain noninvasive measures of blood glucose levels such an approach in humans is difficult to achieve and there is still no device approved for the clinical practice. The first time that glucose was detected in blood using NIR spectroscopy was reported in 1989 by Zeller et al. [54]. Posterior studies showed that very small changes in glucose levels created high errors on the prediction of real value [55], so mathematical algorithms of quality control were developed with improved and corrected models [56]. The devices developed so far encountered serious problems due to the uncertainty resulting from the signal to noise ratio (SNR), the signal to interference ratio (SIR) and the wavelengths in the spectrometric measurements in the NIR band. Many systems for noninvasive measurement of blood glucose levels use a window in the NIR band in the range 6500 - 5500 cm-1 (i.e. 1.54 - 1.82 microns) [57]. These wavelengths correspond to the first overtones of the molecular bonds of CH and OH. This window requires path lengths of 5 mm for the reliable measurement of glucose levels in the range of milli-moles (mmol) with an SNR of 15 dB. This level of sensitivity (SNR = 15 dB) is increased dramatically when reflectance techniques are used as opposed to the absorbance techniques detailed above. In fact, by moving slightly to the transmission window range [0.7 1.33] microns, the sensitivity threshold will be approximately 30 dB (as an example requires 30 dBm transmitting power to detect glucose levels by a masked 0dBm noise, without taking into account the effect of dispersal or interference). More generally, spectrometric methods for measuring glucose levels through the skin suffer from these problems (sensitivity / SNR) [58]: dispersion of the transmitted light, heterogeneous distribution of the absorbed light and variability of the structures (analytes) that cause the scattering of light (due to vascular changes and blood

Current Diabetes Reviews, 2012, Vol. 8, No. 1

51

oxygen saturation SPO2, ignorance of the way of light is propagated inside the tissue, (i.e diffuse spread), heterogeneous distribution of glucose (interstitial, intracellular, blood, etc. ...), the presence of other light absorbing components, which interfere with the measurements (e.g., water, hemoglobin, tryglicerids), similarities between the absorption spectrum of water and glucose, and temperature dependence. One possibility in order to minimize the noise associated with the bone, muscles, other tissues or body fluids (i.e. sweat) is to use the optical signals from pulsatile microcirculation. Yamakoshy showed in 2006 very good results in healthy individuals [59]. Chen et al. in 2008, analyzed the variations in the spectrum of light transmitted through the fingers of 15 diabetic patients, based on the signals from pulsatile microcirculation [60]. The device was promoted under the name of TANGTEST. The measurements were very accurate when done in a stable environment, with 100% of the values situated in the A and B zones of the Clarke´s Error Grid, when compared with a classical glucometer (not with a gold standard method). Nevertheless, the study was realized under special conditions in order to minimize any possible influence on microcirculation, such as prohibiting any physical activity, and maintaining a stable heart rate and a stable environmental temperature 10 minutes prior to measurement. Also, the position of the finger was very important, no changes being allowed prior to or during measurement. The environmental temperature was found to be very important. The patients were tested in conditions of changing temperature, showing errors not present at a normal environmental temperature, suggesting that microcirculation in diabetic patients is worse than non diabetic patients and that non-invasive measurement is not very reliable. Patients with DM had a reduction in the vascular compliance [61-65]. Woolam et al. [66] conducted a comparative study of 52 diabetic patients with 87 normal controls (ages 4 to 75 years) and found a reduction in distensibility of the vessel walls in all young diabetic patients. These observations were confirmed by the study of Lehman et al. [67]. In addition, McVeigh et col. [68] demonstrated that vascular compliance is significantly reduced in diabetic patients regardless of the presence / absence of complications. This alteration in vascular compliance appears more related to connective tissue disorders and / or the accumulation of advanced glycation end products (AGEs) than with increasing thickness of vessel walls [69-71]. Most of the devices showed a very poor correlation between the estimated and the real value of blood glucose during hypoglycemia and in conditions of rapid changes of the glucose concentration [72, 73]. One possible hypothesis for this constant problem with almost all of the devices based on NIR spectroscopy could be that by depending on the signal of pulsatile microcirculation, they really measure the glucose inside the red blood cell. The relationship between red blood cell glucose and plasma glucose in humans is not known, but in vitro studies have shown a rapid equilibration between a stable glucose level outside and inside the red blood cell in 5 minute time, via the GLUT 1 transporter. The glucose level inside the cell was stable for at least 1 hour,

52 Current Diabetes Reviews, 2012, Vol. 8, No. 1

and then the sugar seemed to be drawn out of the cell via an unknown mechanism [74]. This is an interesting concept, but data is lacking regarding such behavior in vivo and in conditions of rapid changes in blood glucose levels. Nevertheless, Gabriely et al. reported a very good correlation between the estimated and the real value of blood glucose, in conditions of rapid changes induced by insulin, with a 98,8% of the estimates values in the A and B zones of the Clarke´s Error Grid. This experiment was performed on 10 healthy individuals and 2 diabetic patients [75]. 3. REAL PROBLEMS IN DIABETIC PATIENTS AND FUTURE PERSPECTIVES In spite of some encouraging results, no device based on non-invasive glucose measurement is currently available in clinical practice for diabetic patients. As mentioned above, the diabetic population has specific characteristics which make the application of non-invasive glucose monitoring techniques difficult in real life. Vascular compliance is significantly reduced in diabetic patients regardless of the presence / absence of complications. This alteration in vascular compliance appears more related to connective tissue disorders and / or accumulation of advanced glycation end products (AGEs) than with the increasing thickness of vessel walls. Type 2 diabetic or glucose intolerant patients have differences in cutaneous blood flow [71] altered peripheral vasomotion and impaired response to temperature changes [76, 77]. Also, acute hyperglycemia affects endothelium dependent vasodilatation, and acute hyperinsulinemia induces acute vasodilatation in humans [78, 79]. The vasodilatory effect of hyperinsulinemia is well demonstrated in healthy subjects and in type 1 diabetic patients [79, 80]. A sudden rise and or fall in glucose and/or insulin levels will lead to changes in the vascular bed. Blood perfusion defects can confound the correlation between NIR signal change and glucose concentration. The light reflected by the skin of a diabetic patient has a different pattern, depending on the metabolic control and the degree of local protein glycation. Another limitation might be that in diabetic patients there are changes to the refractive index and the membrane structure of the red blood cells [81,82]. All these differences between a diabetic patient and a healthy subject could have limited until now the development of glucose non-invasive monitoring devices. But efforts are now being made to overcome these limitations. Recently, Caduff et al. reported a multisensory system for non-invasive glucose monitoring [83]. This interesting concept was reported first by the same group in 2009 [72]. It consists basically of combining several different sensors, with the intention to measure the direct perturbing effects of glucose changes together with the glucose related effects on the tissues. This method would allow for more reliable blood glucose estimation, as compared to a single sensor device. The results obtained in 6 type 1 diabetic patients showed a very promising correlation above 0.8, and 87, 2-89% of all

Ciudin et al.

the results were situated in the A and B zones of the Clarke Error Grid. Nevertheless, the company developing the multisensory has ceased the operations. Harman-Boehm et al. used the same concept to develop the GlucoTrack device, combining three NI methods: ultrasonic, electromagnetic and thermal. The results showed a good correlation when compared with a classical glucometer. It should be noted that the blood glucose values showed in the Clarke´s error grid were above the hypoglycemic range [84]. Another very recent method of non-invasive glucose monitoring, totally different as a concept from those reviewed so far in this paper was reported by Minh et al. [85]. Blood glucose levels were estimated based on the determination of two groups of gases in exhaled breath (acetone+methyl nitrate+ethanol+ethylbenzene and nitrate+propane+methanol and acetone), and showed strong positive correlations (0,883 and 0,869 respectively) in 300 measurements in healthy and type 1 diabetic patients. It should be remarked that the accuracy of the glycemic prediction in the hypoglycemic state was not tested. It seems that by not depending on microcirculation, skin and temperature characteristics these methods could by-pass the problems encountered by the other devices tested so far. Further development of this method will be needed in order to asses its degree of predictive accuracy in the case of hypoglycemia and to evaluate the time lag for some exhaled gases. CONCLUSIONS The frequency of self-determination of glucose levels allows for a tighter control of the disease and slows the progression of the complications. NI glucose determination might promote more frequent testing, but in spite of the efforts made in the past 30-40 years there is no device available at present for use in the clinical practice. Finding a really NI method to determine blood glucose levels in vivo represents a very challenging goal since the fulfillment of the conditions of ideal device is very difficult. Possible explanations could be the combination between the specific features of each method and the specific characteristics of diabetic patients, which makes them respond differently to several physical and chemical methods when compared to their healthy counterparts. Further studies and new techniques development are needed in order to shed more light on this very challenging goal of NI glucose monitoring. REFERENCES [1]

[2]

[3]

The Diabetes Control and Complications Trial Research Group. The Effect of Intensive Treatment of Diabetes on the Development and Progression of Long-Term Complications in Insulin-Dependent Diabetes Mellitus. N Engl J Med 1993; 329: 977-86. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998; 352(9131): 854-65. Erratum in Lancet 1998; 352(9139): 1558. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998; 352(9131): 837-53. Erratum in Lancet 1999; 354(9178): 602.

Non-Invasive Methods of Glucose Measurement [4] [5]

[6] [7]

[8]

[9]

[10] [11]

[12]

[13] [14]

[15]

[16]

[17]

[18] [19]

[20]

[21] [22]

[23] [24]

[25]

Standards of medical care in diabetes. Diabetes Care 2011; 34: S11-S61. Rabinovitch B, March WF, Adams RL. Noninvasive glucose monitoring of the aqueous Humor of the eye: part I. Measurement of very small optical rotations. Diabetes Care 1982; 5: 254-8. Heinemann L, Koschinky T. Continuous glucose monitoring: an overview of today´s technologies and their clinical applications. Int J Clin Pract Suppl 2002; 129: 75-9. Koschinsky T, Jungheim K, Heinemann L. Glucose sensors and alternate site testing-like phenomenon: relationship between rapid blood glucose changes and glucose sensor signals. Diabetes Technol Ther 2003; 5: 829-42. Rebrin K, Steil GM, van Antwerp WP, Mastrototaro JJ. Subcutaneous glucose predicts plasma glucose independent of insulin: implications for continuous monitoring. Am J Physiol 1999; 277: E561-71. Maruo K, Tsurugi M, Tamura M, Ozaki Y. In vivo noninvasive measurement of blood glucose by near-infrared diffuse-reflectance spectroscopy. Appl Spectrosc 2003; 57: 1236-44. Heise HM, Marbach R. Human oral mucosa studies with varying blood glucose concentration by non-invasive ATR-FT-IRspectroscopy. Cell Mol Biol (Noisy-le-grand) 1998; 44: 899-12. Khalil OS. Non-invasive glucose measurement technologies: an update from 1999 to the dawn of the new millennium. Diabetes Technol Ther 2004; 6(5): 660-97. Heinemann L, Kramer U, Klotzer HM, et al. Non-invasive glucose measurement by monitoring of scattering coefficient during oral glucose tolerance tests. Non-Invasive Task Force, Diabetes Technol Ther 2000; 2(2): 211-20. Yeh SJ, Hanna CF, Khalil OS. Monitoring blood glucosevchanges in cutaneous tissue by temperature-modulated localized reflectance measurement. Clin Chem 2003; 49: 924-34. Wickramasinghe Y, Yang Y, Spencer SA. Current problems and potential techniques in in-vivo glucose monitoring. J Fluorescence 2004; 14: 513-20. Waynant RW, Chenault VM. Overview of Non-Invasive Optical Glucose Monitoring Techniques; Overview of non-invasive fluid glucose measurement using optical techniques to maintain glucose control in diabetes mellitus. LEOS Newsletter 1998; Volume12, Number 2. Avialable at: http://photonicssociety.org/newsletters/apr98/overview.htm [Accessed on: 22/3/2206]. Shen Y, Lu Z, Spiers S, MacKenzie HA, Ashton HS, Hanningan J. Measurement of the optical absorption coefficient of a liquid by use of a time-resolved photoacustic technique. Appl Opt 2000; 39: 4007-12. Weiss RAM, Yegorchikov Y, Shusterman A, Raz I. Noninvasive continous glucose monitoring using photoacustic technologyresults from the first 62 subjects. Diabetes Technol Ther 2007; 9(1): 68-4. Ermolina I, Polevaya Y, Feldman Y. Analysis of dielectric spectra of eukaryotic cells by computer modeling. Eur Biophys J 2000; 29: 141-5. Polevaya Y, Ermolina I, Schlesinger M, Ginzburg BZ, Feldman Y. Time domain dielectric spectroscopy study of human cells. Normal and malignant white blood cells. Biochim Biophys Acta 1999; 1419: 257-71. Alavi SM, Gourzi M, Rouane A, Nadi M. Proceedings of the 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Istanbul, Turkey, 2003. Caduff AM, Hirt E, Feldman Y, Ali Z, Heinemann L. First human experiments with a novel non-invasive, non-optical continuous glucose monitoring system. Biosens Bioelectron 2003; 19: 209-17. Gourzi M, Rouane A, Guelaz R, Nadi M, Jaspard F. Study of a new electromagnetic sensor for glycaemia measurement: in vitro results on blood pig. J Med Eng Technol 2003; 27: 276-81. Park JH, Kim CS, Choi BC, Ham KY. The correlation of the complex dielectric constant and blood glucose at low frequency. Biosens Bioelectron 2003; 19: 321-24. Wentholt IME, Hoekstra JBL, Zwart A, DeVries JH. Pendra goes Dutch: lessons for the CE mark in Europe. Diabetologia 2005; 48: 1055-58. Forst T, Caduff A, Talary M, et al. Impact of environmental temperature on skin thickness and microvascular blood flow in diabetic and non-diabetic subjects. Diabetes Technol Therap 2006; 81(1): 94-101.

Current Diabetes Reviews, 2012, Vol. 8, No. 1 [26]

[27]

[28]

[29]

[30] [31]

[32] [33]

[34] [35] [36] [37]

[38] [39]

[40] [41]

[42] [43]

[44] [45]

[46] [47] [48]

[49]

53

Gourzi M, Rouane A, Guelaz R, Alavi MS, McHugh MB, Nadi M. Non-invasive glycaemia blood measurement by electromagnetic sensor: study in static and dynamic blood circulation. J Med Eng Technol 2005; 29: 22-6. Moran GR, Jeffrey KR, Thomas JM, Stevens JR. A dielectric analysis of liquid and glassy solid glucose/water solutions. Carbohydr Res 2000; 328: 573-84. Gebhart S, Faupel M, Fowler R, et al. Glucose sensing in transdermal body fluid collected under continuous vacuum pressure via micropores in the stratum corneum. Diabetes Technol Ther 2003; 5(2): 159-66. Kurnik RT, Oliver JJ, Waterhouse SR, Dunn T, Jayalakshmi Y, Lesho M. Application of the mixtures of experts algorithm for signal processing in a noninvasive glucose monitoring system. Sens Actuators B Chem 1999; 60: 19-26. Pitzer KR, Desai S, Dunn T, Edelman S, Jayalakshmi Y, Kenedy J. Detection of hypoglycemia with the GlucoWatch biorapher. Diabetes Care 2001; 24: 881-5. The diabetes research in children network study group (DIRECNET). Accuracy of the GlucoWatchG2 Biographer and the Continuous Glucose Monitoring System During Hypoglycemia Experience of the Diabetes Research in Children Network. Diabetes Care 2004; 27: 722-6. Domschke A, March WF, Kabilan S, Lowe C. Initial clinical testing of a holographic non-invasive contact lens glucose sensor. Diabetes Technol Ther 2006; 1: 89-93. Badugu R, Lakowicz JR, Geddes CD. A glucose-sensing contact lens: from bench top to patient. Curr Opin Biotechol 2005; 16: 1007. Lane JD, Krumholz DM, Sack RA, Morris C. Tear glucose dynamics in diabetes mellitus. Curr Eye Res 2006; 11: 895-01. Badugu R, Lakowicz JR, Geddes CD. A glucose sensing contact lens: a non-invasive technique for continuous physiological glucose monitoring. J Fluores 2003; 13: 371-4. Geddes CD, Badugu R, Lakowicz JR. Contact lenses may provide window to blood glucose. Biophotonics Int 2004; 2: 50-4. Chu MX, Miyajima K, Takahashi D, et al. Soft contact lens biosensor for in situ monitoring of tear glucose as non-invasive blood sugar assessment. Talanta 2011; 83(3): 960-5. Yeh SJ, Hanna CF, Khalil OS. Monitoring blood glucose changes in cutaneous tissue by temperature-modulated localized reflectance measurements. Clin Chem 2003; 49: 924-34. Sandby-Møller J, Poulsen T, Wulf HC. Influence of epidermal thickness, pigmentation and redness on skin autofluorescence. Photochem Photobiol 2003; 77: 616-20. Hanlon EB, Manoharan R, Koo T-W, Shafer KE, Motz JT. Prospects for in vivo Raman spectroscopy. Phys Med Biol 2000; 45:R1R59. Steffes PG. Laser-based measurement of glucose in the ocular aqueous humor: an efficacious portal for determination of serum glucose levels. Diabetes Technol Ther 1999; 1(2): 129-33. Borchert MS, Storrie-Lombardi MC, Lambert JL. A noninvasive glucose monitor: preliminary results in rabbits. Diabetes Technol Ther 1999; (2): 145-51. Caspers PJ, Lucassen GW, Puppels GJ. Combined in vivo confocal Raman spectroscopy and confocal microscopy of human skin. Biophys J 2003; 85: 572-80. Lipson J, Bernhardt J, Block U, et al. Requirements for calibration in noninvasive glucose monitoring by Raman spectroscopy. J Diabetes Sci Technol 2009; 3(2): 233-41. Roychoudhury P, Harvey LM, McNeil B. At-line monitoring of ammonium, glucose, methyl oleate and biomass in a complex antibiotic fermentation process using attenuated total reflectance-midinfrared (ATR-MIR) spectroscopy. Anal Chim Acta 2006; 561: 218-24. Khalil OS. Spectroscopic and clinical aspects of non-invasive glucose measurements. Clin Chem 1999; 45: 165-77. Arnold MA. Non-invasive monitoring of metabolites using near read infrarred spectroscopy: state of the art. Horm Metab Res 1996; 7: 46-9. Chung H, Arnold MA, Rhiel M, Murhammer DW. Simultaneous measurements of glucose, glutamine, ammonia, lactate and glutamate in aqueous solutions by near-infrared spectroscopy. Appl Spectrosc 1996; 50: 270-6. Riley MR, Arnold MA, Murhammer DW. Adaptive calibration scheme for quantification of nutrients and byproducts in insect cell

54 Current Diabetes Reviews, 2012, Vol. 8, No. 1

[50]

[51] [52]

[53] [54] [55]

[56]

[57] [58]

[59]

[60]

[61]

[62] [63]

[64] [65]

[66] [67] [68]

Ciudin et al.

bioreactors by near infrared spectroscopy. Biotechnol Prog 1988; 14: 527-33. Arnold MA, Small GW. Determination of physiological levels of glucose in an aqueous matrix with digitally filtered Fourier transform near-infrared spectra. Anal Chem 1990; 62: 1457-64. Lu G, Zhou X, Arnold MA, Small GW. Multivariate calibration models based on the direct analysis of near-infrared single beam spectra. Appl Spectrosc 1997; 51: 1330-9. McShane MJ, Coté GL, Spiegelman C. Variable selection in multivariate calibration of spectroscopic glucose sensor. Appl Spectrosc 1997; 51: 1559-64. Riley MR, Arnold MA, Murhammer DW. Effect of sample complexity on quantification of analytes in aqueous samples by near infrared spectroscopy. Appl Spectrosc 2000; 54: 255-61. Zeller H, Novak P, Landgraf R. Blood glucose measurement by infrared spectroscopy. Int J Artif Organs 1989; 12(2): 129-35. Fischbacher C, Jagemann KU, Danzer K, Muller UA, Papenkordt L, Schuler J. Enhancing calibration models for non-invasive nearinfrared specroscopical blood glucose determination. J Anal Chem 1997; 359: 78-82. Heise HM, Bittner A, Marbach R. Near-infrared reflectance spectroscopy for non-invasive monitoring of metabolites. Clin Chem Lab Med 2000; 38(2): 137-45. Hazen KH, Arnold MA, Small GW. Measurement of glucose in water with first-overtone near-infrared spectra. Appl Spectrosc 1998; 52: 1597-05. Heinmann L, Schmelzeisen-Redeker G. On behalf of of the Noninvasive task force (NITF), Non-invasive continuous glucose monitoring in Type I diabetic patients with optical glucose sensors. Diabetologia 1998; 41: 848-54. Yamakoshi K, Yamakoshi Y. Pulse glucometry: A new approach for noninvasive blood glucose measurement using instantaneous differential near-infrared spectrophotometry. J Biomed Opt 2006; 11(5): 054028. Chen CS, Wang KK, Jan MY, et al. Noninvasive blood glucose monitoring using the optical signal of pulsatile microcirculation: a pilot study in subjects with diabetes. J Diabetes Complications 2008; 22(6): 371-6. Pillsbury HC, Hung W, Kyle MC, Freis ED. Arterial pulse waves and velocity and systolic time intervals in diabetic subjects. Am Heart J 1974; 87: 783-90. Megnien JL, Simon A, Valensi P. Comparative effects of diabetes mellitus and hypertension on physical properties of human large arteries. J Am Coll Cardiol 1992; 20: 1562-8. Christensen T, Nubauer B. Increased arterial wall stiffness and thickness in medium-sized arteries in patients with insulindependent diabetes mellitus. Acta Radiol 1988; 29: 299-02. Christensen T and Neubauer B. Arterial wall stiffness in insulindependent diabetes mellitus: an in vivo study. Acta Radiol 1987; 28: 207-8. Thordarson H, Thorgeirsson G, Helgason T. Aortic stiffness in insulin-dependent diabetics: an echocardiographic study. Diabetes Med 1986; 3: 449-54. Woolam GL, Schnur PL, Vallbona C, Hoff HE. The pulse wave velocity as an early indicator of atherosclerosis in diabetic subjects. Circulation 1962; 25: 533-9. Lehmann ED, Gosling RG, Sonksen PH. Arterial wall compliance in diabetes, Diabetes Med 1992; 9: 114-9. McVeigh G, Brennan G, Hayes R. Vascular abnormalities in noninsulin dependent diabetes mellitus identified by arterial waveform analysis. Am J Med 1993; 95: 424-30.

Received: July 09, 2011

[69]

[70] [71]

[72]

[73]

[74]

[75]

[76]

[77]

[78] [79] [80]

[81] [82]

[83]

[84]

[85]

Oxlund H, Rasmussen LM, Andreassen TT, Heickendorf L. Increased aortic stiffness in patients with type I (insulin decedent) diabetes mellitus. Diabetologia 1989; 32: 748-52. Airaksinen, KE, Salmela PI, Linnaluoto MK. Diminished arterial elasticity in diabetes: association with fluorescent advance glycosylation end products in collagen, Cardiovasc Res 1993; 27: 942-45. Salomaa V, Riley W, Kark JD. Non-insulin-dependent diabetes mellitus and fasting glucose and insulin concentrations are associated with arterial stiffness indexes, the ARIC study. Circulation 1995; 91: 1432-43. Caduff A, Talary M, Mueller M, et al. Non-invasive glucose monitoring in patients with type 1 diabetes: a multisensory system combining sensors for dielectric and optical characterization of skin. Biosens Bioelectron 2009; 24: 2778-84. Müller UA, Mertes B, Fischbaucher C, Jageman KU, Danzer K. Non-invasive blood glucose monitoring by means of near infrared spectroscopy: methods for improving the reliability of the calibration models. Int J Artif Organs 1997; 20: 285-90. Khera PK, Joiner CH, Carruthers A, et al. Evidence for interindividual heterogeneity in the glucose gradient across the human red blood cell membrane and its relationship to hemoglobin glycation. Diabetes 2008; 57: 2445-52. Gabriely I, Wozniak R, Mevorach M, Kaplan J, Aharon K, Shamoon H. Transcutaneous glucose measurement using nearinfrared spectroscopy during hypoglycemia. Diabetes Care 1999; 22: 2026-32. Stansberry KB, Shapiro SA, Hill MA, McNitt PM, Meyer MD, Vinik AI. Impaired peripheral vasomotion in diabetes. Diabetes Care 2006; 19: 715-21. Haak E, Haak T, Grozinger Y, Krebs G, Usadel KH, Kusterer K. The impact of contralateral cooling of skin capillary blood cell velocity in patients with diabetes mellitus. J Vasc Res 1998; 35: 24549. Williams SB, Goldfine AB, Timimi FK, et al. Acute hyperglycemia attenuates endothelium-dependent vasodilation in humans in vivo. Circulation 1998; 97: 1695-01. Yki-Jarvinen H, Utriainen T. Insulin-induced vasodilatation: physiology or pharmacology. Diabetologia 1998; 41: 369-79. Oomen PHN, Kant GD, Dullaart RPF, Reitsma WD, Smit AJ. Acute hyperglycemia and hyperinsulinemia enhance vasodilatation in Type-1 diabetes mellitus without increasing capillary permeability and inducing endothelial dysfunction. Microvasc Res 2002; 63: 1-9. Mazarevica G, Freivalds T, Jurka A. Properties of erythrocyte light refraction in diabetic patients. J Biomed Opt 2002; 7: 244-47. Fusman R, Rotstein R, Elishkewich K, et al. Image analysis for the detection of increased erythrocyte, leukocyte and platelet adhesiveness/aggregation in the peripheral blood of patients with diabetes mellitus. Acta Diabetol 2001; 38: 129-34. Caduff A, Mueller M, Megej A, et al. Characteristics of a multisensor system for non invasive glucose monitoring with external validation and prospective evaluation. Biosens Bioelectron 2011, [Epub ahead of print]. Harman-Boehm I, Gal A, Raykhman AM, Naidis E, Mayzel Y. Noninvasive glucose monitoring: increasing accuracy by combination of multi-technology and multi-sensors. J Diabetes Sci Technol 2010; 4(3): 583-95. Minh TD, Oliver SR, Ngo J, et al. Non-invasive measurement of plasma glucose from exhaled breath in healthy and type 1 diabetic subjects. Am J Physiol Endocrinol Metab 2011 300(6): E1166-75. Epub 2011 Apr 5.

Revised: August 30, 2011

Accepted: September 09, 2011