Technology behind commercial devices for blood glucose monitoring in diabetes management: A review

Page 1

R

Tm

SFa

b

c

d

e

f

a

ARRAA

KBEGIG

NT

0d

Analytica Chimica Acta 703 (2011) 124– 136

Contents lists available at ScienceDirect

Analytica Chimica Acta

jou rn al hom epa ge: www.elsev ier .com/ locate /aca

eview

echnology behind commercial devices for blood glucose monitoring in diabetesanagement: A review

andeep Kumar Vashista,b,1, Dan Zhenga,c,1, Khalid Al-Rubeaand, John H.T. Luonge,f,wu-Shan Sheua,b,∗

NUSNNI-NanoCore, National University of Singapore, T-Lab Level 11, 5A Engineering Drive 1, Singapore 117580, SingaporeDepartment of Electrical and Computer Engineering, National University of Singapore, Engineering Drive 1, Singapore 117576, SingaporeDepartment of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, SingaporeUniversity Diabetes Center, King Saud University, P.O. Box 18397, Riyadh 11415, Saudi ArabiaBiotechnology Research Institute, National Research Council Canada, Montreal, Quebec, Canada H4P 2R2Department of Chemistry, University College Cork, Cork, Ireland

r t i c l e i n f o

rticle history:eceived 11 April 2011eceived in revised form 11 July 2011ccepted 13 July 2011vailable online 23 July 2011

a b s t r a c t

The blood glucose monitoring devices (BGMDs) are an integral part of diabetes management now-a-days.They have evolved tremendously within the last four decades in terms of miniaturization, rapid response,greater specificity, simplicity, minute sample requirement, painless sample uptake, sophisticated soft-ware and data management. This article aims to review the developments in the technologies behindcommercial BGMD, especially those in the areas of chemistries, mediators and other components. The

eywords:lood glucose monitoring deviceslectron mediatorslucose limiting membranes

technology concerns, on-going developments and future trends in blood glucose monitoring (BGM) arealso discussed.

© 2011 Elsevier B.V. All rights reserved.

nterferenceslucose strips

Sandeep Kumar Vashist is a Ph.D. in Biotechnologyfrom the Nanotechnology Division, Central ScientificInstruments Organisation, Chandigarh, India. He hasworked extensively in disease diagnostics, biosen-sors, assay development, signal enhancement, rapidimmunoassays and diabetic biomarkers. Before join-ing the NUS-KSU Diabetic Research Partnership atNUSNNI-NanoCore, National University of Singaporeas Team Leader in 2009, he was a bioanalytical scien-tist at Bristol-Myers Squibb Company, Dublin, Ireland.His present research interests comprise the devel-opment of transdermal glucose sensor and insulindrug delivery system for diabetic monitoring andmanagement.

∗ Corresponding author at: Department of Electrical and Computer Engineering,ational University of Singapore, Engineering Drive 1, Singapore 117576, Singapore.el.: +65 65162857.

E-mail address: [email protected] (F.-S. Sheu).1 These authors contributed equally.

003-2670/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.aca.2011.07.024

Dan Zheng received her M.Sc. in Applied Chemistryfrom South China University of Technology in 2009and is presently pursuing her Ph.D. in Chemistry atNational University of Singapore under the NUS-KSUDiabetes Research project on the development oftransdermal glucose sensor and insulin drug deliv-ery for diabetics. Her main research interests are thedevelopment of novel electrochemical strategies fortransdermal glucose sensing using nanotechnology-based approaches.

Khalid Al-Rubeaan is the founder and director of theUniversity Diabetes Center. He is a fellow of Royal Col-lege of Physician and Surgeons of Canada with specialcompetence on endocrinology and metabolism since1989. He is currently holding a post of an assistantprofessor of medicine and a consultant endocrinolo-gist in the Medical College at King Saud University. Heis keenly interested in diabetes clinical practices.

Page 2

Chimi

1

papllrtnitoirdnFbtaok

eb(Ptpmot2abU2id

itmoo1bA

employs a two- or three-electrode system composed of working,reference and auxiliary electrodes. The sensing layer, containingactive materials such as enzymes that react specifically with the

S.K. Vashist et al. / Analytica

John H.T. Luong is the Head of Nanobiotechnologyand Biosensor Technology at the National ResearchCouncil Canada and a Walton fellow (Science Foun-dation of Ireland), Department of Chemistry atUniversity College Cork, Ireland. He has publishedover 260 research papers and 10 patents. He hasreceived several awards for his contribution to variousresearch areas such as nanobiotechnology, biosensortechnology, separation science, and biochemical engi-neering. He has served as an editorial member andGuest Editor of several journals. He has received over5800 citations with an h-index of 41.

. Introduction

Glucose, a primary source of energy for the body’s cells, is trans-orted from the intestines or liver to cells via the bloodstream. It isvailable for cell absorption via insulin, which is produced by theancreas. The human body naturally tightly regulates blood glucose

evels as a part of metabolic homeostasis. The normal blood glucoseevel is 4–8 mM (72–144 mg dL−1), whereas the pathophysiologicalange is 2–30 mM (36–540 mg dL−1). A persistently high level abovehe normal range is referred to as diabetes mellitus, the most promi-ent disease related to failure of blood sugar regulation. Diabetes is

ncurable but manageable, known as Type I (juvenile onset) owingo the ineffectiveness of the pancreas to produce sufficient insulinr Type II to reflect the inability of the body to use the producednsulin. Keeping the blood glucose level within the physiologicalange is essential for diabetics to lead a healthy lifestyle by avoidingiabetes-associated complications such as kidney damage, blind-ess, neuropathy, effect on circulatory system, amputations, etc.or instance, diabetic retinopathy (damage to the retina) is causedy complications of diabetes mellitus, which can eventually leado blindness. This ocular manifestation of systemic disease couldffect up to 80% of all patients who have had diabetes for 10 yearsr more. Therefore, regular blood glucose monitoring (BGM) is aey requirement for diabetics [1–6].

Indeed, diabetes monitoring and management is taking a heavyconomic toll and financial burden on society. The disease haseen declared as a global epidemic by World Health OrganisationWHO) based on its unprecedented alarmed increase worldwide.resently, there are about 285 million diabetics, which are expectedo increase to 439 million by 2030 with an increase of 7 millioneople developing diabetes each year [7,8]. There are about 3.96illion deaths per year attributable to diabetes, which is about 7%

f the mortality in the age group of 20–79 years. About 11.6% ofhe total global healthcare expenditure was spent on diabetes in010. The estimated healthcare expenditure spent in 2010 to treatnd prevent diabetes and associated complications was US$ 376illion, and is expected to increase to US$ 490 billion by 2030 [9].nited Nations passed the landmark resolution on December 20,006 to recognize diabetes as a serious global health concern that

mposes a heavy economic burden. November 14 has since beenesignated as the World Diabetes Day.

Self-glucose monitoring should be used in patients on intensivensulin therapy at least three times daily. Thus, there have been con-inuously increasing research efforts in the area of blood glucose

onitoring as shown by the tremendous increase in the numberf articles published in the previous decades (Fig. 1). Several tensf thousands peer-reviewed articles have been published, of which6,000 have been published just in the last 3 years. Over half of theiosensors produced worldwide are employed for glucose sensing.bout 10 billion glucose assays are performed worldwide every

ca Acta 703 (2011) 124– 136 125

Fwu-Shan Sheu is currently a Principle Investigatorand Chair for Nanobiotechnology research group inthe NanoCore of the National University of Singapore.He establishes a laboratory of using nanomaterialsfor developing various bio- and chemical sensors formedical applications. In the year of 1999, he receivedthe prestigious Oversea Young Scholar ResearchAward in the life science area from the National Nat-ural Science Foundation of China outstanding fromcompetition of all tertiary Institutes of Hong Kong.In 2008, he received an award of the Science andTechnology Pioneer from the Sino-Singapore SuzhouIndustrial Park co-developed by China and Singapore

Government in recognition of his leadership in a techno-entrepreneur team forinnovative nanotechnology.

year. The market of blood glucose meters is highly lucrative as itaccounts for 85% of the total biosensors market. It is projected toreach US$ 6.1 billion by 2012 [10] from the estimated market of US$5 billion by Newman et al. in 2004 [11]. The largest market of glu-cose meters is in the United States, where it is expected to reach US$1.28 billion by 2012. The market is dominated by a small numberof large diagnostic companies such as Abbott, Roche Diagnostics,Bayer, Minimed and LifeScan. The performance of the commercialBGMS from these companies has been extensively studied recently[12–19].

The tremendous commercial potential has always been themajor stimulant for the continuous development of technologiesbehind commercial BGMD, which has evolved at a very fast pacein the last few decades and ultimately led to better diabetes man-agement. The essential prerequisites of successful BGMD are theirsimple operating procedure, user-friendliness, simple presenta-tion, greater accuracy and reliability, low cost, enhanced memoryand sophisticated software-based data management. This reviewarticle will highlight the major developments in the technologybehind commercial BGMD, especially in the areas of chemistries,glucose limiting membranes and anti-interference techniques. Thefuture trend and technical challenges of blood glucose monitoringare also discussed with an emphasis on commercial feasibility.

BGMD is based on an electrochemical sensor, which is cost-effective with rapid response and capable for mass productionalong with convenient set-up. A typical electrochemical sensor

Fig. 1. Number of articles published during the mentioned period pertaining toblood glucose monitoring.Data was taken on April 01, 2011 from www.sciencedirect.com using “blood glucosemonitoring” in the advanced search option.

Page 3

1 Chimi

gfiaidiacea

2

BcdftgsdLBmMtiinTstttHme

otsnai

d1mfiomiOtmde(

cme5e

26 S.K. Vashist et al. / Analytica

lucose molecules, is coated on the working electrode. This con-guration allows the working electrode potential to be measuredgainst the reference electrode without compromising the stabil-ty of the reference electrode by passing current over it. The analyteiffuses into the sensor through a porous membrane to the work-

ng electrode, where it is oxidized or reduced, thereby generatingn electric current, which then passes through the external circuitomprising of amplifiers and other signal processing devices. Thelectrical signal is then converted to the analyte detection signalnd displayed.

. History

Canadian surgeon Frederick Banting and his assistant Charlesest developed the analytical methods for the determination of glu-ose in blood and urine in the early 20th century. Clark and Lyonsescribed the glucose biosensor in 1962, which initiated the questor the glucose monitoring technology from the Children’s Hospi-al of Cincinnati, Ohio, USA [20]. Updike and Hicks immobilizedlucose oxidase (GOx) in a gel on an oxygen electrode to mea-ure glucose in biological fluids [21]. The Ames Reflectance metereveloped by Clemens in 1971 [22] was the first BGMD by Milesaboratories, USA (now part of Bayer). In comparison to currentGMD, it was a bulky and expensive device. Yellow Springs Instru-ent Company, USA launched the glucose analyser (known asodel 23A YSI analyzer) in 1975 based on the amperometric detec-

ion of hydrogen peroxide, which is employed till date as a standardn clinical diagnosis. The company has continuously developedmproved models of glucose analysers although the basic mecha-ism remains the same. GOx is retained between two membranes.he outer polycarbonate membrane prevents large interfering sub-tances to come inside, retains the GOx inside but allows glucoseo pass through. The inner cellulose acetate membrane allowshe hydrogen peroxide produced by the glucose reaction to passhrough to the platinum electrode for amperometric detection.owever, the technology cannot be miniaturized to make glucoseeters. The instrument is robust but expensive due to the platinum

lectrode, which is almost exclusively used in clinical laboratories.The screen-printing technique was adapted for the production

f amperometric biosensors in 2004 [23] and employed by allhe major companies for the bulk-manufacture of glucose sensingtrips for many electrochemical glucose sensing devices. The tech-ology has developed a lot in recent years with the development of

wide variety of inks based on carbon, metals and other particlesncluding multi-walled carbon nanotubes (MWCNTs) [24].

MediSense, previously known as Genetics International, intro-uced the concept of electrochemical sensing of blood glucose in987. It was purchased by Abbott for US$ 867 million in 1996, whichade Abbott the world leader in the glucose meters’ market. The

rst electrochemical glucose monitor named ExacTech [25], basedn a pen and a strip consisting of GOx and ferrocene-derivativeediator, was introduced by Abbott. It was developed by Genet-

cs International in collaboration with Universities of Cranfield andxford. The group at Cranfield University, UK contributed towards

he strip development and reported the development of its ampero-etric glucose monitor [25]. There had been continuous technology

evelopments in the glucose monitors from 1987 onwards by sev-ral companies resulting in the development of various productsTable 1).

The design of the glucose meters changed from pen-shape toompact mobile-shape; the sample requirement has been mini-

ized to 300 nL; the memory for storing glucose readings has been

nhanced; the glucose sensing duration has been minimized to s; and, the data port and PC-based software management hasnabled sophisticated data analysis for healthcare personnel for

ca Acta 703 (2011) 124– 136

better medical treatments. Although the technology has reachedthe maturation stage, but with the continuously increasing glucosesensing markets in the developing nations and due to the non-availability of diabetic cure, the market position of the companieswill still be dependent on the continuous technology develop-ments. The in vivo continuous glucose monitoring and closed-loopsystems are the new trends in the glucose sensing market althoughtheir market is very limited.

The dominance in the glucose sensing market is the major causeof mergers and acquisitions for the leading companies. Roche Diag-nostics was formed in 1998 by the merger of Roche with BoehringerMannheim, a German company with an extensive biosensor port-folio and a strong market position in diagnostics. In 1987, LifeScanintroduced the test strip design, made of a flat plastic piece with ahole that was covered by a membrane, in its One Touch® system.LifeScan also acquired the BGM technology from Inverness MedicalTechnology for US$ 1.3 billion stock-for-stock transaction in 2001.Abbott was losing their market share to TheraSense Inc., cofoundedby Adam Heller and his son Ephraim Heller in 1996. StrategicallyAbbott purchased TheraSense Inc. in 2004 for US$ 1.26 billion tomaintain its dominant position in the BGM market. TheraSenseInc. became a part of Abbott Diabetes Care (ADC). The companyintroduced a thin-layer microcoulometer utilizing 300 nL of bloodsample (the lowest ever sample volume used by any glucose meter)in the Abbott’s FreeStyleTM BGM in 1999. ADC introduced FreeStyleNavigator, a continuous glucose monitoring system (CGMS) basedon a wired-enzyme concept, in 2007. Adam Heller was awarded theU.S. National Medal of Technology and Innovation on September 29,2008 for his significant contribution to the field of electrochemistryand technology development for diabetic BGM.

3. Technologies behind BGMD

3.1. Enzymes

Enzymes, belonging to the family of oxidoreductases, are used inall commercial BGM strips for the highly specific glucose detection.The most popular enzymes employed are GOx and glucose dehy-drogenase (GDH). GOx used in the commercial BGMD is mostlyisolated from Aspergillus niger. It is a dimeric protein of 160 kDawith each monomer composed of an identical polypeptide chain. Astrongly bound redox cofactor, flavin adenine dinucleotide (FAD),is located at the reactive site of each subunit. It accepts electronsfrom glucose and is oxidized by oxidizing substances such as dioxy-gen (O2). During glucose oxidation, the oxidized form (FAD–GOx)first reacts with glucose (Eq. (1)) followed by the oxidation ofthe reduced form (FADH2–GOx) and the production of hydrogenperoxide (H2O2) (Eq. (2)). Hydrogen peroxide is oxidized at a cat-alytic, classically platinum (Pt) anode as H2O2 → 2H+ + O2 + 2e−

with the electron flow proportional to the number of blood glu-cose molecules. Stable GOx is commercially available at low costand withstands greater extremes of operating conditions, i.e. lessstringent conditions during the manufacturing process. The directelectron transfer between the GOx active site and the surface of aconventional electrode is limited due to a thick protein layer, whichsurrounds the FAD redox center and results in an intrinsic barrier.Therefore, natural or artificial mediators are required to re-oxidizethe GOx–FADH2).

GDH belongs to the class of quinoproteins, which use pyrrolo-quinoline quinone (PQQ) as cofactor to convert glucose togluconolactone [26,27]. GDH is also a dimeric enzyme composed

of two identical protein monomers with each monomer binding aPQQ molecule and three calcium ions [28]. One of the three cal-cium ions activates the PQQ cofactor, whereas the other two arerequired for the functional dimerization of the GDH molecule. The

Page 4

S.K.

Vashist

et al.

/ A

nalytica Chim

ica A

cta 703 (2011) 124– 136

127

Table 1Characteristics of commercial BGMS.

Company Products Detection method Enzyme Sample Sampling site Sample size (�L) Test time (s) Range(mg dL−1)

Abbott FreeStyle Navigator® Amperometric (CGMS),coulometric (BGMS)

WIREDENZYMETM

technology

Whole blood, capillary Upper arm, forearm, hand,fingertips, thigh and calf

0.3 (BGMS) 7 20–500

FreeStyle Freedom® Lite Coulometric GDH Whole blood, capillary Upper arm, forearm, hand,fingertips, thigh and calf

0.3 5 20–500

FreeStyle® Lite Coulometric GDH Whole blood, capillary Upper arm, forearm, hand,fingertips, thigh and calf

0.3 5 20–500

Precision XtraTM Amperometric GDH Whole blood, capillary Upper arm, forearm, hand,fingertips, thigh and calf

0.6 5 20–500

MediSense® OptiumTM

XceedTMAmperometric GDH Whole blood, capillary Fingertips, forearm, upper arm,

the base of the thumb0.6 5 20–500

Minimed Guardian® Amperometric GOx Whole blood, capillary CGMS Continuous Real-time 40–400

Dexcom Seven® Amperometric GOx Whole blood, capillary CGMS Continuous Real-time 40–400SEVEN® PLUS Amperometric GOx Whole blood, capillary CGMS Continuous Real-time 40–400

Roche Diagnostics ACCU-CHEK® Compact Plus Reflectance photometric GDH Whole blood, capillary Fingertips, palm, forearm,upper arm, thigh and calf

1.5 5 10–600

ACCU-CHEK® Active Reflectance photometric GDH Whole blood, capillary Fingertips 1-2 5 10–600ACCU-CHEK® Aviva Amperometric GDH Whole blood, capillary Fingertips, palm, forearm,

upper arm, thigh and calf0.6 5 10–600

ACCU-CHEK® AdvantageACCU-CHEK® CompactACCU-CHEK® Complete

Bayer BREEZE® 2 Amperometric GOx Whole blood, capillary Fingertips, palm and forearm 1 5 20–600Contour® Amperometric GDH Whole blood Fingertips, palm and forearm 0.6 5 20–600Ascensia EliteTM Amperometric GOx Whole blood Finger, alternative puncture

site within certain conditions2 30 20–600

Ascensia EliteTM XL Amperometric GOx Whole blood Finger, alternative puncturesite within certain conditions

2 30 20–600

Lifescan One Touch® Ultra® 2 Amperometric GOx Whole blood, capillary Fingertip, forearm and palm 1 5 20–600One Touch® Ultra LinkTM Amperometric GOx Whole blood, capillary Fingertip, forearm and palm 1 5 20–600One Touch® UltraMiniTM Amperometric GOx Whole blood, capillary Fingertip, forearm and palm 1 5 20–600One Touch® SelectTM Amperometric GOx Whole blood, capillary Fingertip, forearm and palm 1 5 20–600One Touch® UltraSmart® Amperometric GOx Whole blood, capillary Fingertip, forearm and palm 1 5 20–600

Page 5

1 Chimica Acta 703 (2011) 124– 136

ot(

G

F

F

P

peiTFict1

GOwBagctnth

hcssccrmgcn(amicx

3

etotrefFrCr

Fig. 2. Chemical structure of Os2+/3+ complex that was used as mediator in the (A)1st [67], (B) 2nd and (C) 3rd generation of Abbott Freestyle test strips [82].

28 S.K. Vashist et al. / Analytica

xidation mechanism of glucose by PQQ-dependent GDH is similaro that of FAD–GOx [29] with the exception that the reduced formPQQH2–GDH) is not oxidized by O2 [30,31] (Eqs. (1)–(4)).

lucose + FAD–GOx → ı − Gluconolactone + FADH2–GOx (1)

ADH2–GOx + O2 → FAD–GOx + H2O2 (2)

AD + 2H+ + 2e− → FADH2 (3)

QQ + 2H+ + 2e− → PQQH2 (4)

The apparent reduction potential of FAD–GOx at 25 ◦C in thehysiological medium at pH 7.2 is −0.048 V vs. standard hydrogenlectrode (SHE) [32], whereas the reduction potential of PQQ–GDHs 10.5 ± 4 mV vs. SHE at pH 7.0 in the presence of excess Ca2+ [33].he two enzymes also differ in their specificity for glucose [34,35].AD–GOx is highly specific to glucose [36], although mannose cannterfere even at low concentration [37]. PQQ–GDH exhibits similaratalytic efficiency to both glucose and maltose [38]. The electronransfer turnover rates of FAD–GOx and PQQ–GDH are 5000 s−1 and1,800 s−1, respectively at 35 ◦C [39].

Nicotinamide adenine dinucleotide (NAD)- and FAD-dependentDH have also been used in commercial BGM strips. Abbott’sptiumTM blood glucose electrodes employ the NAD–GDH system,hereas Bayer’s ContourTM TS test strips use the FAD–GDH system.oth NAD– and FAD–GDH systems are quite specific for glucosepart from being independent of O2. Xylose may interfere with thelucose detection of NAD–GDH [40]. In addition, FAD–GDH mayonvert other non-glucose sugars such as maltose, mannose, galac-ose and lactose but only to a very small extent [41]. Although theseon-glucose sugars are not present in diabetics or healthy persons,hey may be present in individuals taking specific medication oraving a rare disease condition.

The US Food and Drug Administration (FDA) agency’s publicealth notification in August, 2009 [42] stated that the GDH–PQQhemistry can cause potentially fatal errors in the glucose mea-urements in patients on medications that contain non-glucoseugars. Non-glucose sugars such as maltose, xylose and galactosean falsely elevate the glucose results, which would suggest medi-al intervention leading to an inappropriate insulin dose that mayesult in hyperglycemia, coma, or death. Therefore, FDA recom-ended the public and healthcare facilities to avoid GDH–PQQ

lucose test strips, which were used by many commercial glu-ose meters. Patients on the following medications containingon-glucose sugars should not use such strips. They are extranealicodextrin) peritoneal dialysis solution, orencia (abatacept), adeptdhesion reduction solution (4% icodextrin), BEXXAR radioim-unotherapy agent, octagram 5%, WinRho SDF liquid, vaccinia

mmune globulin intravenous (human), HepaGamB and productsontaining or metabolized into non-glucose sugars such as maltose,ylose, and galactose.

.2. Mediators

A mediator is a small organic or inorganic chemical capable ofxisting in both oxidized and reduced forms, which reacts quicklyo donate or receive electrons [43]. In commercial BGMS, whenxidoreductases oxidize glucose to gluconolactone, electrons areransferred from glucose to the oxidized form of mediator therebyeducing it. The reduced mediator is then re-oxidized by thelectrode for electrochemical detection. The O2/H2O2 redox pairunctions as a mediator to transfer electrons in the case of the

AD–GOx system using O2 as an oxidizing agent. The oxidation ofeduction product, i.e. H2O2 is shown in Eq. (5). The commercialGM systems using this technology are GlucoWatch® G2 Biog-apher (Cygnus, Redwood City, CA, withdrawn from commercial

Page 6

S.K. Vashist et al. / Analytica Chimica Acta 703 (2011) 124– 136 129

e pho

dC

H

iflcmdgttF

F

2

rrteecHct

fBhaaa�pr

Fig. 3. The mediator system used in Roch

istribution), Guardian REAL-Time (Medtronic MiniMed, Sylmar,A) and DexComTM STSTM-7 (DexCom, Inc., San Diego, CA), etc. [43].

2O2 → 2H+ + O2 + 2e− (5)

The use of dioxygen, however, can cause serious device lim-tation known as “oxygen deficit” [44], mainly attributed to theuctuations in oxygen tension and the insufficient oxygen con-entration compared to that of glucose in interstitial fluid, whichay be several hundred folds greater [43]. This limitation intro-

uces variability in the sensor response and decreases the upperlucose detection limit. The use of a glucose limiting membrane inhe BGM strips can circumvent this problem. An alternative media-or for O2 can also be used to facilitate the redox reaction betweenAD–/FADH2–GOx and Mred/Mox as shown in Eq. (6).

ADH2–GOx + 2Mox → FAD–GOx + 2Mred + 2H+ (6)

Mred → 2Mox + 2e− (7)

The mediator should be able to compete with O2 and to reactapidly with enzyme cofactors at low redox potential. The loweredox potential is necessary to avoid interference from other elec-roactive biomolecules such as bilirubin, uric acid and ascorbic acid,tc., which introduce inaccuracy in glucose readings. As mentionedarlier, the normal physiological level of glucose is from 4 to 8 mMompared to 0.1 mM for ascorbic acid and 4-acetamidophenol.owever, these two species at a platinum electrode generate higherurrents, larger than that of highly concentrated glucose, owing toheir very high electron transfer rates.

Ferrocene-derivatives [45–47], Os2+/3+ complexes [48,49] anderricyanide are the most commonly used mediators in commercialGM strips. First reported in 1984 [50], ferrocene and its derivativesave been demonstrated as very efficient electron transfer medi-tors for the enzymatic reaction catalyzed by GOx [51–55]. Theyre small-size molecules, which enable them to approach to the

ctive site of GOx [56–58] and thereby provide strong overlap of the-orbitals of the cyclopentadienyl ring with those of the enzymerosthetic group. 1,1′-dimethyl-3-(2-amino-1-hydroxyethyl) fer-ocene was used as a redox mediator in MediSense® ExacTechTM

tometric and electrochemical strips [72].

and Precision QIDTM BGMS. Os2+/3+ complexes are widely employedas mediators for glucose sensing devices as they also can transferelectrons rapidly between the redox centers of enzymes and theelectrode surface [59–66]. The Os polypyridine-based complexeshave been employed in the Abbott Freestyle test strips. The Osredox mediator (Fig. 2(A)) [67] was used in the PQQ–GDH basedFreestyle test strips, which reduced the assay time to <15 s at a rel-atively low potential of −125 mV vs. Ag/AgCl. This redox mediatorwas improved further to a very fast mediator (Fig. 2(B)), whichreduced the assay time to 5 s. Abbott then shifted the PQQ–GDHto FAD–GDH system and further improved its redox mediator(Fig. 2(C)), thereby leading to higher current response at the redoxpotential of −160 mV vs. Ag/AgCl with an identical assay time of 5 s.Ferricyanide has also been used as a fast mediator for the enzymaticreaction [68–71]. It is a widely used mediator in the present com-mercial BGMS such as Bayer ContourTM BGM strips and LifeScanOneTouch® Ultra® BGM strips as the mediator reaction of hexa-cyanoferrate III/hexacyanoferrate II is very simple.

Nitrosoaniline has been employed as the precursor of media-tor in Roche’s photometric and electrochemical BGM strips (Fig. 3)[72]. The mediator is produced by the reaction of nitrosoanilinewith enzyme (e.g. GOx, PQQ– and FDA–GDH) and glucose. Themediator cycle starts from the compound quinine diimine oxide,which is oxidized by enzyme and glucose to generate a very activeintermediate, i.e. hydroxylamine whose amino group is a strongelectron donor. Therefore, hydroxylamine decays easily to quininediimine and electrons are transferred rapidly to a working electrodeor an indicator (used in photometric glucose meter) without anyenzyme. The reverse reaction, i.e. the reduction of nitrosoanilinecan be realized in an electrochemical BGMS by applying a reductionpotential.

The next generation of BGMD would be devoid of mediators andhave low applied potential close to that of enzyme’s redox poten-tial in order to improve the detection selectivity for glucose. In this

case, electrons are transferred directly from glucose to the elec-trode surface via the active centers of enzymes. Nevertheless, asdiscussed previously, the spatial separation of the donor–acceptorpair is a critical challenge for these devices.

Page 7

130 S.K. Vashist et al. / Analytica Chimi

Fl

3

eepwmesrtfGTaei

molecules to react with the immobilized enzyme molecules while

ig. 4. Chemical structure of (A) redox hydrogels [80,81], (B) cross-linker, (C) epoxyinkage for Wired EnzymeTM [43].

.3. Wired enzyme technique

The wired enzyme technique solved the leaching of solublenzymes, which is the major limitation of CGMS. The leaching ofnzymes is responsible for the shorter lifespan of products and irre-roducibility in results. Interestingly, the concept of wired enzymesas initially devised to improve the leaching of diffusional redoxediators during the shuttling of electrons between enzyme and

lectrode surface. The first enzyme wiring experiment [73] demon-trated that GOx could be modified with ferrocene/ferriciniumedox centers, which were able to form phonon-assisted elec-ron tunneling paths by the formation of ferrocene carboxamidesrom the reaction of amine groups of inner protein domains ofOx and the carbodiimide-activated ferrocene/ferricinium ester.hese paths transfer the electrons between the electrode surface

nd the FAD/FADH2 redox centers buried inside GOx with greaterfficiency. Therefore, glucose oxidation occurs without any leach-ng of the redox centers. The usage of the electron-conducting

ca Acta 703 (2011) 124– 136

hydrogels in glucose sensing pioneered by Heller and his collab-orators [73–78] proved that GOx could be covalently bound toelectron-conducting hydrogels in a leach-proof manner by theelectrical wiring of its redox centers. In brief, GOx is tethered toan insoluble, but water-swollen, cross-linked polymer-networkof the gel [49,79]. Since enzymes are enveloped in the redoxhydrogels, their active centers can be electrically connected toelectrodes regardless of the spatial orientation of enzymes. Thehydrogels can also connect multiple enzyme layers so that 10-foldor in some cases 100-fold higher current densities are accessiblewhen most of the redox centers are electrically connected to theelectrode surfaces, in comparison to those obtained by enzymemonolayers. For instance, the current densities of glucose electro-oxidation can exceed 1 mA cm−2 at 0–100 mV versus Ag/AgCl[49,79–81]. The electron-conducting hydrogels are composed of Ospolypyridine-based redox centers that are tethered to the cross-linked water-soluble polymer backbones [43,82]. Fig. 4(A) shows a“wire” formed by vinyl pyridine polymer with pendant Os complex[80,81]. The shown redox hydrogels do not change their struc-ture upon redox reaction and thus have low Marcus reorganizationenergy [83], enabling rapid electron transmission by the redox cen-ters. Fig. 4(B) shows a cross-linker, which connects the exposedamine groups on GOx to the polymer backbone’s pyridyl groups viaepoxy linkages (Fig. 4(C)) and simultaneously cross-links the poly-mer chains to form an insoluble hydrogel on the electrode surface[84]. The mechanism for electrocatalysis of glucose oxidation bythe wired GOx can be described as follows: initially an electrostaticadduct of GOx, a polyanion at pH 7.3, is formed with an excess ofa polycationic redox polymer; it is then crosslinked to avoid phaseseparation during the GOx incorporation in the Os2+/3+ complex-based redox polymers [82]. In the water swollen hydrogels, theredox polymers conduct electrons or holes through self-exchange[85] resulting from Marcus-type collisional electron transfer [86],i.e. phonon-assisted tunneling. Here the FAD centers are reducedto FADH2 during glucose oxidation by the electrically wired GOxin the redox hydrogel, after which the FADH2 centers collide withthe Os3+ centers and then electrons/holes transfer occurs. Finally,Os3+ is reduced in the hydrogel to Os2+, which can be re-oxidizedto Os3+. Since the electron transfer by self-exchange requires colli-sions between reduced and oxidized redox centers [85,87], electrondiffusion is optimum when reduced and oxidized centers are aboutequal, i.e. when the hydrogel is poised at its redox potential. The rateof the self-exchange of electrons/holes is fastest when the redoxfunctions are tethered to the polymer network by long and flexiblespacers (optimally 10–15 atom long) which can increase the ampli-tude of the displacement of the tethered redox centers and therebyenable efficient electron-transferring collision [80,81]. For exam-ple, the apparent electron diffusion coefficient of the cross-linkedredox copolymer shown in Fig. 4(A) is 6 × 10−6 cm2 s−1, compara-ble to the freely diffusing Os2+/3+ redox pair. The enzyme-wiringredox hydrogels arrange the GOx molecules in a manner that thedistance between the electrode surface and the FAD/FADH2 redoxcenters in GOx can be minimized. They are highly permeable toglucose, gluconolactone and the electrolytes, and provide three-dimensional electrocatalysts [83,88] with a high current densityfor glucose detection [81] to allow the miniaturization of glucosetesting electrodes [89–91].

3.4. Glucose limiting membrane

A glucose limiting membrane is essential for BGMD using theO2/H2O2 redox pair as a mediator. It limits the excess glucose

maximizing the availability of O2 [92]. The Guardian REAL-TimeCGMS (Medtronic MiniMed, Sylmar, CA) employs a proprietarypolyurethane polyurea block copolymer, which is composed of

Page 8

S.K. Vashist et al. / Analytica Chimica Acta 703 (2011) 124– 136 131

F by (B) hexamethylene diisocyanate, (C) aminopropyl-terminated siloxane polymer and(

hp(wctdtCpr[pc

BoibdDcacemwfni

3

aatodlg(

ig. 5. Chemical structures of (A) polyurethane polyurea block copolymer formedD) polyethylene glycol [43].

examethylene diisocyanate, aminopropyl-terminated siloxaneolymer and polyethylene glycol, as glucose limiting membraneFig. 5) [43]. The hydrophobic siloxane is highly permeable to O2,hereas the hydrophilic diol provides facile permeability to glu-

ose molecules. An optimum balance can be obtained between theransmission of O2 and glucose by altering the ratio of siloxane toiol. Such elaborately synthesized copolymer is ideal to get rid ofhe “oxygen deficit” problem. The DexComTM STSTM-7 CGMS (Dex-om, Inc., San Diego, CA) also employs a proprietary polyurethaneolymer as its glucose limiting membrane, which minimizes theequired O2 concentration and thus, reduces the H2O2 generated93]. This design is important for CGMS using the O2/H2O2 redoxair as mediator because H2O2 is a strong oxidizing substance thatan damage the enzyme activity and even the device.

A glucose limiting membrane is necessary for commercialGMS to counteract the “oxygen deficit” and lower concentrationf immobilized enzymes in comparison to the high physiolog-cal glucose concentration. Therefore, the immobilized enzymesecome saturated at a particular glucose concentration and will notetect the higher concentrations. FreeStyle Navigator CGMS (Abbottiabetes Care, Alameda, CA) employs a vinyl pyridine–styreneopolymer (Fig. 6(A)) with an epoxy cross-linker (Fig. 6(B))s its glucose limiting hydrogel membrane [94]. The polymeran be functionalized to introduce desirable properties such asnhanced hydration and biocompatibility. The functionalized poly-er hydrates immediately after insertion in the patient’s skin,hereas the polymer backbone hydrates in several hours. The

unctionalized hydrogel-coated sensor is highly biocompatible aso encapsulation was observed on the sensor for one year after

mplantation in rabbit’s muscle.

.5. Anti-interference technique

The biological and analytical interfering substances such ascetaminophen [95], ascorbic acid, bilirubin, cholesterol, dopaminend uric acid can react with the immobilized enzymes, and affecthe accuracy and reliability of the BGM strips. The interferencesf ascorbic acid, acetaminophen and dopamine with the glucose

etection was studied using commercial BGMS [96]. At low glucose

evels, low concentration of ascorbic acid (30 �g mL−1) affects thelucose reading of Accu-Chek Advantage (Roche) and One TouchLifescan), whereas Accu-Chek Advantage could also be affected by

Fig. 6. Chemical structures of (A) vinyl pyridine-styrene copolymer and (B) epoxycross-linker used in FreeStyle Navigator CGMS [43].

low levels of acetaminophen and dopamine (20 �g mL−1). At highglucose level, 40 �g mL−1 of dopamine affects the glucose read-

−1

ing of Accu-Chek Advantage, while 20 �g mL of acetaminophenaffects the glucose reading of Precision G and Precision QID(Abbott). There is a wide range of electrochemical interfering sub-stances and medications that need to be tested for analyzing the

Page 9

1 Chimi

pttD

risislbpaOpemritat

4

dampmwpsiiaciAidnc

esBibie

eish(

G

A

G

i

32 S.K. Vashist et al. / Analytica

otential interferences in diabetics as they can interfere with a par-icular BGMD. These include creatinine, bilirubin (unconjugated),riglycerides, ibuprofen, tetracycline, salicylic acid, ephedrine, l-OPA, methyl-DOPA, tolbutamide and tolazamide.

A simple mediator such as ferricyanide has minor kinetic bar-iers, which results in its higher likelihood to react with thenterfering substances along with glucose [72]. Therefore, a moreuitable mediator with higher chemical selectivity is desired ast will have a very low probability of reacting with interferingubstances. An alternative approach is the use of an interferenceimiting membrane, which can oxidize the oxidizable interferentsy chemical reactions. For instance, lactate oxidase and horseradisheroxidase were used in an earlier version of FreeStyle Navigators anti-interference ingredients. Lactate oxidase was oxidized by2 resulting in the generation of H2O2 that oxidized horseradisheroxidase, which then reacted with the active interferents. How-ver, the requirement of dissolved O2 is the major drawback of thisethod. Therefore, the redox potential of the enzyme-catalyzed

eaction has to be reduced to eliminate the oxidation reactions ofnterferents. In the current version of the FreeStyle BGM meter withhe Os2+/3+ complex as a redox mediator for the wired enzyme, thepplied potential is reduced to −160 mV, which is much less thanhe oxidation potentials of most interferents vs Ag/AgCl.

. BGM strip design

A commercial BGM strip should have high accuracy, high repro-ucibility, rapid assay time, smaller blood sample requirementnd greater stability. It should be cost-effective and capable ofass-manufacture with high reproducibility. Another important

arameter in the design of BGM strips is the reproducibility perfor-ance. The common components included in a BGM strip are theorking, counter and reference electrodes that are deposited on alastic substrate. In order to confirm the sufficient filling of bloodamples in the strips, fill detection electrodes are also employedn most commercial BGM strips. The automatic fill detection elim-nates the visual confirmation error and reduces the testing times the electrochemical assay starts immediately after the strip isompletely filled with the blood sample. A small capillary chambers located on the electrode substrate to work as reaction container.

mixture of enzymes, mediators and other chemical componentss coated within the capillary chamber in dry form. Fig. 7 shows theesigns of the Abbott’s FreeStyle BGM test strip [97], the subcuta-eous wired GOx electrode [91] and the FreeStyle Navigator sensorhip [82].

A working electrode is the channel to transmit glucose-derivedlectrons from blood sample to the meter. It is generally made fromcreen-printed carbon ink or vapor-deposited gold or palladium. InGM strips, the area of the working electrode is kept constant dur-

ng mass-production to ensure high reproducibility. The distanceetween the working and the auxiliary/reference electrode is min-

mized to decrease the required blood volume and inter-electrodelectrolytic resistance.

In most commercial BGM strips, both auxiliary and referencelectrodes are usually combined. The most frequently used auxil-ary/reference electrode is the Ag/AgCl electrode, which is made bycreen printed Ag/AgCl ink using polyester as a binding agent. Thealf-cell reactions and the net electrode reaction are shown by Eqs.8), (9) and (10), respectively. W.E. denotes the working electrode.

lucose → ı − Gluconolactone + 2H+ + 2e− (W.E.) (8)

gCl + e− → Ag + Cl− (9)

lucose + 2AgCl → ı − Gluconolactone + 2Ag + 2H+ (10)

Besides the Ag/AgCl electrode, an inert conductor auxil-ary/reference electrode, made of the material of the working

ca Acta 703 (2011) 124– 136

electrode, is also employed so that the working and auxil-iary/reference electrodes can be simultaneously deposited on thesubstrate.

The working and auxiliary/reference electrodes have beenassembled by a coplanar and facing electrode configuration. Thefacing electrode configuration is preferred in comparison to thecoplanar one as the distance between the working and the aux-iliary/reference electrodes can be reduced, which results in lessersample requirement and faster assay time. However, if the distancebetween the electrodes is too close, it can induce “redox shut-tling” due to the shuttling of large amounts of mediator back andforth between the working and counter/reference electrode underhigh potential difference (ca. 400–500 mV) during glucose oxida-tion. Therefore, fast enzyme–mediator chemistry with low appliedpotential is desired to avoid redox shuttling. The Os2+/3+ complexused in Abbott FreeStyle BGMD makes the “redox shuttling” ther-modynamically impossible because it has an oxidation potentialnegative than that of the Ag/AgCl counter electrode, but positivethan that of glucose-reduced enzyme. The oxidized Os3+ complexcannot be obviously reduced by the more positively poised Ag/AgCl[82]. This led to significant reduction in the assay time from 45 s to5 s at the applied potential of −160 mV vs Ag/AgCl.

The capillary chamber is actually very small having the capac-ity of about 1 �L. It has an entrance at one end to receive bloodsample and another opening to let the displaced air escape as thesample fills into the strip. The strip filling time of the capillarychamber is directly proportional to the square of chamber lengthand inversely proportional to capillary thickness and cosine wet-ting angle, according to the Washburn equation (Eq. (11)) [98]:

t = 3�x2

[� cos(�w)]s(11)

where t is filling time, � is viscosity, x is length along fill axis,� is liquid surface tension, �w is wetting angle and s is capillarythickness.

Therefore, in order to develop a rapid device, the filling timecan be efficiently reduced by decreasing the chamber length andby reducing the wetting angle by surfactant treatment.

The reagents used in a BGM strip are the enzymes, redox medi-ators, enzyme stabilizer, film forming agents (for glucose limiting,anti-interference and anti-biofouling membrane) and others suchas filling time reducing surfactants and biocompatible interface.In finger stick BGM strips, a mixture of enzyme and mediator isusually coated over the working electrode in aqueous form, whichleads to the formation of an active reaction film after evaporation.Sometimes these reagents are initially mixed with the carbon inkand then co-deposited on the substrate. However, this manufac-turing method is not suitable for implantable CGMS as the reagentsmay leach off and dissolve in the subcutaneous fluid. Therefore, theenzyme and the mediator are covalently immobilized in a poly-meric membrane in case of CGMS [99]. The design and technologybehind CGMS is described elsewhere [100,101].

The coulometric detection method was devised by Abbott fora tiny blood sample, i.e. 300 nL [79] as such a submicroliter sam-ple can be taken in a painless manner. The developed coulometricBGMD had improved linearity between the blood glucose con-centration and the detected electrical signal in comparison to theprevious devices. The comparison of Abbott’s coulometric BGMD

with those from other companies (Table 1) shows that coulometricdevices requires the least sample although they have the same testtime, and a smaller glucose detection range than BGMD from RocheDiagnostics, Bayer and Lifescan.

Page 10

S.K. Vashist et al. / Analytica Chimica Acta 703 (2011) 124– 136 133

bcuta

5

if

Fig. 7. The expanded views of Abbott’s (A) FreeStyle BGM test strip [97], (B) su

. Enzymeless approach

The direct electrochemical oxidation of glucose at neutral pHs a very complex process involving adsorption, electron trans-er, and subsequent chemical rearrangement. Platinum, one of the

neous wired GOx electrode [91], and (C) FreeStyle Navigator sensor chip [82].

best electrode materials, is still unable to provide adequate sen-

sitivity to glucose. Other drawbacks include poor selectivity andsurface foulings affected by chloride ion, amino acids, creatinine,epinephrine, urea, ascorbic acid, and uric acid in blood. Therefore,the platinum loses its sensitivity to glucose and the linear range for

Page 11

134 S.K. Vashist et al. / Analytica Chimica Acta 703 (2011) 124– 136

Table 2Comparison of recent commercially available glucose meters.

Features Abbott FreeStyle® Lite Bayer ContourTM USB LifeScan OneTouch®

Ultra® 2Roche DiagnosticsAccu-Chek® Aviva

Sample volume (�L) 0.3 0.6 1 0.6Test time (s) 5 5 5 5Insufficient sample Yes Yes May get ERROR 5 or an

inaccurate resultYes

2nd sample application Permits redosing within 60 s – Not allowed Permits redosing within 5 sCalibration Plasma equivalent Plasma equivalent Plasma equivalent Plasma equivalentPrecision (%) CV ≤ 3 CV ≤ 5.3 CV ≤ 4.4 CV ≤ 2Hematocrit (%) 15–65 15–65 30–55 20–70Correlation efficient (�) 0.98 0.98 0.98 –Enzymes used GDH GDH GOx GDHMediators Os complex Ferricyanide Ferricyanide Ferricyanide

printi

gowc[wfrOaRsi

6

aadciottbmmdlecmfopBimBsm

amleSv

Enzyme loading Screen-printing Screen-Strip need coding No NoCE/FDA approval Yes Yes

lucose detection becomes very narrow. Electrocatalytic oxidationf glucose can be realized on the glassy carbon electrode modifiedith multi-walled carbon nanotube. Unfortunately, such electro-

atalysis can only be conducted in basic condition at pH over 9102]. The Pt–Pb alloy is capable of oxidizing glucose at neutral pHith larger response compared to pristine platinum. However, sur-

ace poisoning is still problematic as the response signal diminishesapidly to zero when 100 mM NaCl is present in the sample [103].f particular interest is the use of nanoporous platinum (NPt) with

roughness factor (RF) of 72 for the glucose oxidation [104–106].F plays an important role in the oxidation of glucose but is notensitive to AA and AP, and NPt retains high sensitivity to glucosen the presence of 0.15 mM NaCl.

. Conclusions and future trends

Diabetes is a global health concern and is increasing at anlarming rate. The possibility of a cure for diabetes seems unre-listic in the short term although intensive efforts are beingevoted towards the development of artificial pancreas and pan-reatic or islet cell transplantation. However, they may suffer frommmune suppression and unavailability of organs and thus maynly provide a temporary solution. Therefore, BGM is very impor-ant for diabetics to maintain their blood glucose level withinhe physiological range and thus live a healthy life. There haveeen tremendous developments in the technology behind com-ercial BGMD, especially in the areas of chemistries for glucoseeter strips, minimization of sample size, decrease in glucose

etection time, reduction of potential interferences, better glucoseimiting membranes, miniaturization of glucose strips and meters,fficient electrode design, increased user-friendly operating pro-edure, enhanced device memory, intelligent software-based dataanagement and wireless data transmission. The R&D involvement

or these technology developments requires a substantial amountf money. Therefore, the high-end research can only be accom-lished with continuous funding support from a market giant inGMD. There is a highly lucrative market worth billions of dollars

n BGMD, which is the main driving force for technology develop-ents in this area. The market is segmented with Abbott, LifeScan,

ayer, Roche, Minimed and Dexcom as the key players. Table 2hows the comparison of four commercially available blood glucoseeters.Time and considerable efforts are also required to validate the

ccuracy, precision, and reliability of BGMD. The U.S. FDA recom-ends that glucose sensors must have an error of <20% for glucose

evels between 1.65 and 22 mmol L−1 when compared to the refer-nce laboratory measurements. The International Organization fortandardization (ISO) 15197:2003 stipulates that 95% of the indi-idual glucose results must be within ±0.83 mmol L−1 (15 mg dL−1)

ng Screen-printing Screen-printingYes YesYes Yes

for glucose concentrations <4.2 mmol L−1 (<75 mg dL−1) and within±20% at glucose concentrations ≥4.2 mmol L−1 (≥75 mg dL−1). TheISO Technical Committee ISO/TC 212 also released a protocol (theISO 15197 guideline) to validate the accuracy and repeatability ofglucose monitoring devices at three to five different glucose levelsin real-life situations.

There have been continuous incremental improvements in thedevices, softwares and associated features, which provide tremen-dous support for a company to maintain its dominant marketposition. However, there is continuous competition among themarket giants to search for the breakthroughs in technology devel-opments, which can provide a competitive edge to a company andwould enable them to dominate the market. The healthcare facili-ties are growing at a tremendous pace in the developing countriessuch as India and China, which account for a substantial glucosesensing market. Therefore, there is a growing competition amongthe companies to expand their markets.

Future improvements in precision and accuracy seem unneces-sary for the next-generation of BGMD as all the commercial glucosemeters have a very high precision of less than 3% coefficient ofvariation and very less deviation of about 5% from the lab refer-ence method such as conventional hexokinase assay. However, theblood glucose meters have always been extensively investigatedand reviewed by the regulatory and healthcare agencies. This isevidenced by the FDA/Center for Devices and Radiological Healthpublic meeting on blood glucose meters [107] that was held onMarch 16–17, 2010, where the clinical accuracy requirements, per-formance, interferences and limitations of the blood glucose meterswere discussed and reviewed. FDA has advocated for tightening upthe system accuracy criteria standards for the blood glucose metersto enable tight glycaemic control in diabetics.

Considerable efforts have been devoted to the non-invasiveBGM in the last two decades but there is no major success dueto the absence of specific interactions with the glucose molecules.The continuous BGM is another highly prospective area that canprovide highly valuable information to healthcare personnel totake informed decisions for better therapy and may also providevaluable disease insights. However, the difficulty in handling thesedevices and the higher cost may be a limiting factor. The in vivo CGMhas been made possible by Abbott’s FreeStyle Navigator, Medtronic’sGuardian Real-Time and DexCom’s SEVEN PLUS and SEVEN STS.

Many types of software have been made for effective diabetesmanagement. They track the diabetes-related health informationand provide user-friendly information to the patients and health-care professionals in the form of graphs, charts and reports. These

include Abbott’s CoPilot Health Management System, Roche’s Accu-Chek’s 360◦ Diabetes management System, LifeScan’s OneTouchDiabetes Management software, Medtronic’s CareLink Pro TherapyManagement Software for Diabetes, and DexCom’s Seven System

Page 12

Chimi

DwbfecsemwA

fartatbcdwtbgsoelorclcTscafCBmf

oictaadisoosoafltNwwti

S.K. Vashist et al. / Analytica

ata Manager 3. Apart from these, the state-of-the-art mobile soft-are for diabetic monitoring known as GluCoMo [108] is developed

y Artificial Life, Inc. It is the first major telemedicine applicationor the mobile devices for personalized diabetes management. Itnables the diabetics to monitor their blood glucose levels, foodonsumption, insulin dose, activities and other information. Theoftware is based on the Artificial Life’s telematics platform, whichnables diabetics, healthcare professionals and hospitals to com-unicate and exchange information in an extremely use-friendlyay using web portals or mobiles. The software is available at thepple Store for the Apple devices.

Cost-effectiveness and convenience has been the main factorsor the developments in technology till date for BGMD. Almostll commercial BGMD are very simple and easy-to-use, and haveapid response and low sample requirement. Therefore, it seemshat the technologies behind BGMD have already reached thedvanced stage. There are still possibilities for minor incremen-al improvements but it will be highly difficult to come across areakthrough technology that can further evolve the BGMD. Inontrast, there are continuously increasing efforts towards theevelopment of continuous BGMD. But these specialized devicesill only be beneficial to a smaller fragment of the diabetic popula-

ion. A plethora of research is going on in the field of non-invasivelood glucose monitoring using highly diversified technolo-ies [109,110]. These include near infrared spectroscopy, fusionpectroscopy, Raman infrared spectroscopy, light spectroscopy,cclusion optic spectroscopy, mid infrared spectroscopy, thermalmission spectroscopy, fluorescence spectroscopy, microporation,aser spectroscopy, fluorescence, bio-electromagnetic resonance,ptical coherence tomography, radiomolecular magnetism, fluo-escence resonance energy transfer, polarized light, ultrasound,onductivity, heat, electromagnetic radiant ray, bioimpedance,aser crystalline colloidal array, visual pigment bleaching, electro-hemical sweat measurement and metabolic heat conformation.herefore, a number of companies have come up in the glucoseensing market with products based on new glucose sensing con-epts. But it remains to be seen whether these new products will bes accurate and robust as the commercially existing glucose metersrom the dominant companies. A variety of nanomaterials such asNTs and nanosensors are also being employed for BGM [111,112].ut there is a potential concern about the toxicity of these nano-aterials due to the absence of internationally agreed guidelines

or toxicity analysis [113].There will be continuously increasing market for BGMD based

n the increasing diabetic population, absence of diabetic cure andncrease in healthcare facilities and awareness in the developingountries. The next generation of BGMD will be an integrated sys-em that will provide numerous glucose measurements, greaterutomation and software management, advanced lancing mech-nisms, very simple operation, wireless telemetry and insulin drugelivery. Non-invasive glucose analysis is another goal of monitor-

ng. Among optical or transdermal approaches, the optical glucoseensing uses the physical properties of light in the interstitial fluidr the anterior chamber of the eye. The GlucoWatch Biographerf Cygnus (Redwood City, CA, USA), first transdermal glucose sen-or approved by the US FDA, is based on transdermal extractionf interstitial fluid by reverse iontophoresis. It was never widelyccepted in the market and withdrawn in 2008 due to several pit-alls such as inaccuracy, false alarm, skin irritation, sweating, andong warm up time. Although significant efforts have been made inhis area, reliable non-invasive glucose sensing is still not available.umerous nanostructured materials will be developed to pave the

ay for the development of enzymeless glucose sensing platformsith nanoporous platinum as a good example. Nevertheless, sensi-

ivity, reliability, and selectivity of non-enzymatic glucose sensingn human blood remain to be one of the key issues.

ca Acta 703 (2011) 124– 136 135

Acknowledgment

This work was supported by the Research CollaborationAgreement between NUSNNI-NanoCore, National University ofSingapore, Singapore and University Diabetes Center, King SaudUniversity, Saudi Arabia.

References

[1] www.idf.org, 10-05-2011.[2] J.E. Shaw, R.A. Sicree, P.Z. Zimmet, Diabetes Res. Clin. Pract. (2009),

doi:10.1016/j.diabres.2009.10.007.[3] www.diabetesatlas.org, 10-05-2011.[4] American Diabetes Association, Diabetes Care 31 (2008) S12–S54.[5] H. Kolb, K. Kempf, S. Martin, M. Stumvoll, R. Landgraf, Diabetes Res. Clin. Pract.

87 (2010) 150–156.[6] E. Renard, Am. J. Med. 118 (2005) 12S–19S.[7] L. Blonde, A.J. Karter, Am. J. Med. 118 (2005) 20S–26S.[8] R.M. Bergenstal, J.R. Gavin III, Am. J. Med. 118 (2005) 1S–6S.[9] V. Marks, Clin. Chim. Acta 251 (1996) 3–17.

[10] Biosensors in Medical Diagnostics – Global Strategic Business Report,http://www.researchandmarkets.com/reports/338842, 21-05-2011.

[11] J.D. Newman, P.J. Warner, A.P.F. Turner, L.J. Tigwell, Biosensors: A ClearerView, Cranfield University, UK, 2004, p. 216.

[12] A.S. John, W.A. Davis, C.P. Price, T.M.E. Davis, J. Diabetes Complications 24(2010) 129–141.

[13] B. Solnica, J.W. Naskalski, J. Sieradzki, Clin. Chim. Acta 331 (2003) 29–35.[14] C.M. Girardin, C. Huot, M. Gonthier, E. Delvin, Clin. Biochem. 42 (2009)

136–142.[15] S.K. Garg, L.C. Sharp, M.E. Stults, in: V.A. Fonseca (Ed.), Clinical Diabetes:

Translating Research into Practice, China, 2006, pp. 442–457.[16] R. Lehmann, S. Kayrooz, H. Greuter, G.A. Spinas, Diabetes Res. Clin. Pract. 53

(2001) 121–128.[17] R.M. Bergensta, Insulin 3 (2008) 5–14.[18] S. Demir, G.C. Yilmazturk, D. Aslan, Diabetes Res. Clin. Pract. 79 (2008)

400–404.[19] M.M. Kimberly, H.W. Vesper, S.P. Caudill, S.F. Ethridge, E. Archibold, K.H.

Porter, G.L. Myers, Clin. Chim. Acta 364 (2006) 292–297.[20] L.C. Clark Jr., C. Lyons, Ann. N. Y. Acad. Sci. 102 (1962) 29–45.[21] S.J. Updike, G.P. Hicks, Nature 214 (1967) 986–988.[22] A.H. Clemens, Reflectance Meter, US Patent, 14 September 1971, 3,604,815.[23] X.-E. Zhang, in: J.M. Cooper, A.E.G. Cass (Eds.), Screen-Printing Methods for

Biosensor Production, Biosensors: A Practical Approach, 2nd ed., Oxford Uni-versity Press, New York, 2004, pp. 41–58.

[24] J.D. Newman, S.F. White, I.E. Tothill, A.P.F. Turner, Anal. Chem. 67 (1995)4594–4599.

[25] K.O. Kyvik, J. Traulsen, B. Reinholdt, A. Froland, Diabetes Res. Clin. Pract. 10(1990) 85–90.

[26] H.E. Pauly, P. Pfleiderer, Z. Hoppe Seyler’s, Physiol. Chem. 356 (1976)1613–1623.

[27] P. Kratzsch, R. Schmuck, D. Bunk, Z. Shao, D. Thym, W. Knappe, Variants ofSoluble Pyrroloquinoline Quinone-Dependent Glucose Dehydrogenase, WOPatent, 2 May 2002, 34919 A1.

[28] P. Dokter, J. Frank, J.A. Duine, Biochem. J. 239 (1986) 163–167.[29] A. Oubrie, H.J. Rozeboom, K.H. Kalk, A.J.J. Olsthoorn, J.A. Duine, B.W. Dijkstra,

EMBO J. 18 (1999) 5187–5194.[30] J.A. Duine, J. Frank Jr., J.K. Van Zeeland, FEBS Lett. 108 (1979) 443–446.[31] S. Tsujimura, K. Kojima, K. Kano, T. Ikeda, M. Sato, H. Sanada, H. Omura, Biosci.

Biotechnol. Biochem. 70 (2006) 654–659.[32] J. Kulys, L. Tetianec, A. Ziemys, J. Inorg. Biochem. 100 (2006) 1614–1622.[33] A. Sato, K. Takagi, K. Kano, N. Kato, J.A. Duine, T. Ikeda, Biochem. J. 357 (2001)

893–898.[34] T.G. Schleis, Pharmacotherapy 27 (2007) 1313–1321.[35] B. Keilin, H.E. Hartree, Biochem. J. 50 (1952) 331–341.[36] E.C. Adams Jr., R.L. Mast, A.H. Free, Arch. Biochem. Biophys. 91 (1960) 230–234.[37] V. Leskovac, S. Trivic, G. Wohlfahrt, J. Kandrac, D. Pericin, Int. J. Biochem. Cell

Biol. 37 (2005) 731–750.[38] A.J.J. Olsthoorn, J.A. Duine, Biochemistry 37 (1998) 13854–13861.[39] I. Willner, R. Baron, B. Willner, Biosens. Bioelectron. 22 (2007) 1841–1852.[40] B. Ginsberg, J. Diabetes Sci. Technol. 3 (2009) 903–913.[41] Glucose Dehydrogenase FAD Dependent Catalogue No. GLDE-70-1192,

E.C. number 1.1.99.10, http://www.sekisuidiagnostics.com/pdf/FADGDH1192 DATA.pdf, 21-05-2011.

[42] FDA Public Health Notification: Potentially Fatal Errors with GDH-PQQ Glucose Monitoring Technology, http://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/PublicHealthNotifications/ucm176992.htm, 21-05-2011.

[43] J. Hönes, P. Müller, N. Surridge, Diabetes Technol. Ther. 10 (2008) S10–S26.[44] G. McGarraugh, Diabetes Technol. Ther. 11 (2009) S17–S24.[45] J. Wang, Chem. Rev. 108 (2008) 814–825.[46] B. Ballarin, M.C. Cassani, R. Mazzoni, E. Scavetta, D. Tonelli, Biosens. Bioelec-

tron. 22 (2007) 1317–1322.

Page 13

1 Chimi

36 S.K. Vashist et al. / Analytica

[47] V. Laurinavicius, J. Razumiene, A. Ramanavicius, A.D. Ryabov, Biosens. Bio-electron. 20 (2004) 1217–1222.

[48] J. Razumiene, R. Mesˇkys, V. Gureviciene, V. Laurinavicius, M.D. Reshetova,A.D. Ryabov, Electrochem. Commun. 2 (2000) 307–311.

[49] R. Antiochia, L. Gorton, Biosens. Bioelectron. 22 (2007) 2611–2617.[50] L. Ye, M. Hammerle, A.J.J. Olsthoorn, W. Schuhmann, H.-L. Schmidt, J.A. Duine,

A. Heller, Anal. Chem. 65 (1993) 238–241.[51] A.E. Cass, G. Davis, G.D. Francis, H.A.O. Hill, W.J. Aston, I.J. Higgins, E.V. Plotkin,

L.D.L. Scott, A.P.F. Turner, Anal. Chem. 56 (1984) 667–671.[52] S. Viswanathana, J.A. Ho, Biosens. Bioelectron. 22 (2007) 1147–1153.[53] D. Shan, W. Yao, H. Xue, Biosens. Bioelectron. 23 (2007) 432–437.[54] J. Qiu, H. Peng, R. Liang, Electrochem. Commun. 9 (2007) 2734–2738.[55] J.-D. Qiu, M.-Q. Deng, R.-P. Liang, M. Xiong, Sens. Actuators B: Chem. 135

(2008) 181–187.[56] J.-D. Qiu, R. Wang, R.-P. Liang, X.-H. Xia, Biosens. Bioelectron. 24 (2009)

2920–2925.[57] M. Alvarez-Icaza, H.M. Kalisz, H.J. Hecht, K.-D. Aumann, D. Schomburg, R.D.

Schmid, Biosens. Bioelectron. 10 (1995) 735–742.[58] N.J. Forrow, G.S. Sanghera, S.J. Walters, J. Chem. Soc. Dalton Trans. 16 (2002)

3187–3194.[59] N.J. Forrow, S.J. Walters, Biosens. Bioelectron. 19 (2004) 763–770.[60] T.J. Ohara, R. Rajagopalan, A. Heller, Anal. Chem. 66 (1994) 2451–2454.[61] M. Pravda, C.M. Jungar, E.I. Iwuoha, M.R. Smyth, K. Vytras, A. Ivaska, Anal.

Chim. Acta 304 (1995) 127–138.[62] C. Danilowicz, E. Cortoın, F. Battaglini, J. Electroanal. Chem. 445 (1998) 89–94.[63] S. Reiter, K. Habermuller, W. Schuhmann, Sens. Actuators. B: Chem. 79 (2001)

150–156.[64] C. Zhang, Q. Gao, M. Aizawa, Anal. Chim. Acta 426 (2001) 33–41.[65] R. Kurita, H. Tabei, Y. Iwasaki, K. Hayashi, K. Sunagawa, O. Niwa, Biosens.

Bioelectron. 20 (2004) 518–523.[66] P.P. Joshi, S.A. Merchant, Y. Wang, D.W. Schmidtke, Anal. Chem. 77 (2005)

3183–3188.[67] P. Scodeller, V. Flexer, R. Szamocki, E.J. Calvo, N. Tognalli, H. Troiani, A.

Fainstein, J. Am. Chem. Soc. 130 (2008) 12690–12697.[68] C. Taylor, G. Kenausis, I. Katakis, A. Heller, J. Electroanal. Chem. 396 (1995)

511–515.[69] J. Okuda, J. Wakai, K. Sode, Anal. Lett. 35 (2002) 1465–1478.[70] L. Xiang, Z. Zhang, P. Yu, J. Zhang, L. Su, T. Ohsaka, L. Mao, Anal. Chem. 80

(2008) 6587–6593.[71] H. Yao, N. Hu, J. Phys. Chem. B 113 (2009) 16021–16027.[72] J. Biscay, E.C. Rama, M.B.G. García, J.M.P. Carrazón, A.C. García, Electroanalysis

23 (2011) 209–214.[73] Y. Degani, A. Heller, J. Phys. Chem. 91 (1987) 1285–1289.[74] Y. Degani, A. Heller, J. Am. Chem. Soc. 111 (1989) 2358–2361.

[75] M.V. Pishko, I. Katakis, S.E. Lindquist, L. Ye, B.A. Gregg, A. Heller, Angew. Chem.

102 (1990) 109–111.[76] B.A. Gregg, A. Heller, Anal. Chem. 62 (1990) 258–263.[77] B.A. Gregg, A. Heller, J. Phys. Chem. 95 (1991) 5970–5975.[78] B.A. Gregg, A. Heller, J. Phys. Chem. 95 (1991) 5976–5980.

ca Acta 703 (2011) 124– 136

[79] N. Mano, F. Mao, A. Heller, Chem. Commun. 10 (2004) 2116–2117.[80] N. Mano, F. Mao, A. Heller, J. Electroanal. Chem. 574 (2005) 347–357.[81] F. Mao, N. Mano, A. Heller, J. Am. Chem. Soc. 125 (2003) 4951–4957.[82] A. Heller, B. Feldman, Acc. Chem. Res. 43 (2010) 963–973.[83] A. Heller, Curr. Opin. Chem. Biol. 10 (2006) 664–672.[84] B. Feldman, R. Brazg, S. Schwartz, R. Weinstein, Diabetes Technol. Ther. 5

(2003) 769–779.[85] A. Aoki, A. Heller, J. Phys. Chem. 9 (1993) 11014–11019.[86] R.A. Marcus, N. Sutin, Biochim. Biophys. Acta Rev. Bioenergy 811 (1985)

265–322.[87] A. Aoki, R. Rajagopalan, A. Heller, J. Phys. Chem. 99 (1995) 5102–5110.[88] A. Heller, J. Phys. Chem. 96 (1992) 3579–3587.[89] M.V. Pishko, A.C. Michael, A. Heller, Anal. Chem. 63 (1991) 2268–2272.[90] E. Csoregi, C.P. Quinn, D.W. Schmidtke, S.-E. Lindquist, M.V. Pishko, L. Ye, I.

Katakis, J.A. Hubbell, A. Heller, Anal. Chem. 66 (1994) 3131–3138.[91] E. Csoregi, D.W. Schmidtke, A. Heller, Anal. Chem. 67 (1995) 1240–1244.[92] W.P. VanAntwerp, Polyurethane=Polyurea Compositions Containing Silicon

for Biosensor Membranes, US Patent, 16 March 1999, 5,882,494.[93] M.A. Tapsak, R.K. Rhodes, K. Rathbun, M.C. Shults, J.D. McClure, Techniques

to Improve Polyurethane Membranes for Implantable Glucose Sensors, USPatent, 5 June 2007, 7, 226,978.

[94] F. Mao, H. Cho, Biosensor Membranes Composed of Polymers Containing Het-erocyclic Nitrogens, US Patent, 23 August 2005, 6,932,894.

[95] L.J. Cartier, P. Leclerc, M. Pouliot, L. Nadeau, G. Turcotte, B. Fruteau-de-Laclos,Clin. Chem. 44 (1998) 893–894.

[96] Z. Tang, X. Du, R.F. Louie, G.J. Kost, Am. J. Clin. Pathol. 113 (2000) 75–86.[97] B. Feldman, G. McGarraugh, A. Heller, N. Bohannon, J. Skyler, E. DeLeeuw, D.

Clarke, Diabetes Technol. Ther. 2 (2000) 221–229.[98] E.W. Washburn, Phys. Rev. 17 (1921) 374–375.[99] A. Heller, Annu. Rev. Biomed. Eng. 1 (1999) 153–175.

[100] H. Hovorka, Diabet. Med. 23 (2006) 1–12.[101] D.C. Klonoff, Diabetes Technol. Ther. 7 (2005) 770–775.[102] J.S. Ye, Y. Wen, W.D. Zhang, L.M. Gan, G.Q. Xu, F.S. Sheu, Electrochem. Commun.

6 (2004) 66–70.[103] Y. Sun, H. Buck, T.E. Mallouk, Anal. Chem. 73 (2001) 1599–1604.[104] S. Park, T.D. Chung, H.C. Kim, Anal. Chem. 75 (2003) 3046–3049.[105] H. Boo, S. Park, B. Ku, Y. Kim, J.H. Park, H.C. Kim, T.D. Chung, J. Am. Chem. Soc.

126 (2004) 4524–4525.[106] B. Gollas, J.M. Elliott, P.N. Bartlett, Electrochim. Acta 45 (2000) 3711–3724.[107] www.fda.gov/MedicalDevices/NewsEvents/WorkshopsConferences/

ucm187406.htm, 18-05-2011.[108] www.glucomo.com, 18-05-2011.[109] A.M. Tura, G. Pacini, Diabetes Res. Clin. Pract. 77 (2007) 16–40.[110] C.E. Ferrante do Amaral, B. Wolf, Med. Eng. Phys. 30 (2008) 541–549.

[111] S.K. Vashist, D. Zheng, K. Al-Rubeaan, J.H.T. Luong, F.-S. Sheu, Biotechnol. Adv.

29 (2011) 169–188.[112] J.C. Kevin, A.C. Heather, Trends Mol. Med. 16 (2010) 584–593.[113] H.-F. Cui, S.K. Vashist, K. Al-Rubeaan, J.H.T. Luong, F.-S. Sheu, Chem. Res. Tox-

icol. 23 (2010) 1131–1147.