16/08/18

1

AndreaMosca

CentroperlaRiferibilitàmetrologicainMedicinadiLaboratorioDip.Fisiopatologiamedico-chirurgicaedeitrapianti

UniversitàdegliStudidiMilano

Albumina glicata: un nuovo tool a diagnostico a disposizione della clinica Interpretation of HbA1c: analytical and clinical problems

* 

*  references

*  pre-analytical phase

*  analytical phase

*  post-analytical phase

*  future

A.Mosca-UniMI 2

* 

*  references

*  pre-analytical phase

*  analytical phase

*  post-analytical phase

*  future

A.Mosca-UniMI 3

A.M

osca-UniMI

4

Guidelines and Recommendations for Laboratory Analysisin the Diagnosis and Management of Diabetes Mellitus

David B. Sacks,1* Mark Arnold,2 George L. Bakris,3 David E. Bruns,4 Andrea Rita Horvath,5 M. Sue Kirkman,6

Ake Lernmark,7 Boyd E. Metzger,8 and David M. Nathan9

BACKGROUND: Multiple laboratory tests are used to di-agnose and manage patients with diabetes mellitus.The quality of the scientific evidence supporting theuse of these tests varies substantially.

APPROACH: An expert committee compiled evidence-based recommendations for the use of laboratory test-ing for patients with diabetes. A new system was devel-oped to grade the overall quality of the evidence and thestrength of the recommendations. Draft guidelineswere posted on the Internet and presented at the 2007Arnold O. Beckman Conference. The document wasmodified in response to oral and written comments,and a revised draft was posted in 2010 and again mod-ified in response to written comments. The NationalAcademy of Clinical Biochemistry and the EvidenceBased Laboratory Medicine Committee of the AACCjointly reviewed the guidelines, which were acceptedafter revisions by the Professional Practice Committeeand subsequently approved by the Executive Commit-tee of the American Diabetes Association.

CONTENT: In addition to long-standing criteria basedon measurement of plasma glucose, diabetes can bediagnosed by demonstrating increased blood hemo-globin A1c (Hb A1c) concentrations. Monitoring of gly-

cemic control is performed by self-monitoring ofplasma or blood glucose with meters and by laboratoryanalysis of Hb A1c. The potential roles of noninvasiveglucose monitoring, genetic testing, and measurementof autoantibodies, urine albumin, insulin, proinsulin,C-peptide, and other analytes are addressed.

SUMMARY: The guidelines provide specific recommen-dations that are based on published data or derivedfrom expert consensus. Several analytes have minimalclinical value at present, and their measurement is notrecommended.© 2011 American Association for Clinical Chemistry and American

Diabetes Association

Diabetes mellitus is a group of metabolic disorders of car-bohydrate metabolism in which glucose is underutilizedand overproduced, causing hyperglycemia. The disease isclassified into several categories. The revised classifica-tion, published in 1997 (1), is presented in Table 1. Type 1diabetes mellitus, formerly known as insulin-dependentdiabetes mellitus (IDDM)10 or juvenile-onset diabetesmellitus, is usually caused by autoimmune destruction ofthe pancreatic islet beta cells, rendering the pancreas un-able to synthesize and secrete insulin (2). Type 2 diabetes

1 Department of Laboratory Medicine, National Institutes of Health, Bethesda,MD; 2 Department of Chemistry, University of Iowa, Iowa City, IA; 3 Departmentof Medicine, Hypertensive Disease Unit, Section of Endocrinology, Diabetes andMetabolism, University of Chicago, Chicago, IL; 4 Department of Pathology,University of Virginia Medical School, Charlottesville, VA; 5 Screening and TestEvaluation Program, School of Public Health, University of Sydney, SEALSDepartment of Clinical Chemistry, Prince of Wales Hospital, Sydney, Australia;6 American Diabetes Association, Alexandria, VA; 7 Department of ClinicalSciences, Lund University/CRC, Skane University Hospital Malmö, Malmö, Swe-den; 8 Division of Endocrinology, Northwestern University, Feinberg School ofMedicine, Chicago, IL; 9 Massachusetts General Hospital and Harvard MedicalSchool, Diabetes Center, Boston, MA.

* Address correspondence to this author at: National Institutes of Health, De-partment of Laboratory Medicine, 10 Center Dr., Bldg. 10, Rm. 2C306,Bethesda, MD 20892-1508. Fax 301-402-1885; e-mail sacksdb@mail.nih.gov.

Received December 30, 2010; accepted February 28, 2011.Previously published online at DOI: 10.1373/clinchem.2010.16159610 Nonstandard abbreviations: IDDM, insulin-dependent diabetes mellitus;

GDM, gestational diabetes mellitus; FPG, fasting plasma glucose; NHANES,National Health and Nutrition Examination Survey; OGTT, oral glucosetolerance test; NACB, National Academy of Clinical Biochemistry; ADA,American Diabetes Association; GPP, good practice point; IDF, Interna-tional Diabetes Federation; Hb A1c, hemoglobin A1c; QALY, quality-

adjusted life-year; UKPDS, United Kingdom Prospective Diabetes Study;DCCT, Diabetes Control and Complications Trial; CAP, College of AmericanPathologists; DKA, diabetic ketoacidosis; ICU, intensive care unit; SMBG,self-monitoring of blood glucose; GHb, glycated hemoglobin; DiGEM, Di-abetes Glycaemic Education and Monitoring (trial); ISO, InternationalOrganization for Standardization; CGM, continuous glucose monitoring;FDA, US Food and Drug Administration; IADPSG, International Associationof Diabetes and Pregnancy Study Groups; HAPO, Hyperglycemia andAdverse Pregnancy Outcome (study); AcAc, acetoacetate; !HBA,!-hydroxybutyric acid; NGSP, National Glycohemoglobin StandardizationProgram; eAG, estimated average glucose; ADAG, A1c-Derived AverageGlucose (study); ACCORD, Action to Control Cardiovascular Risk in Diabe-tes (study); HEDIS, Healthcare Effectiveness Data and Information Set;MODY, maturity-onset diabetes of the young; ICA, autoantibody to islet cellcytoplasm; HNF, hepatocyte nuclear factor; VNTR, variable nucleotidetandem repeat; IAA, insulin autoantibody; GAD65A, autoantibody to 65-kDa isoform of glutamic acid decarboxylase; IA-2A, autoantibody to insuli-noma antigen 2; IA-2!A, autoantibody to insulinoma antigen 2!; ZnT8A,autoantibody to zinc transporter 8; LADA, latent autoimmune diabetes ofadulthood; DPT-1, Diabetes Prevention Trial of Type 1 Diabetes; DASP,Diabetes Autoantibody Standardization Program; JDF, Juvenile DiabetesFoundation; JNC, Joint National Committee; NKF, National Kidney Foun-dation; eGFR, estimated glomerular filtration rate.

Clinical Chemistry 57:6e1–e47 (2011) Special Report

e1

16/08/18

2

* 

*  references

*  pre-analytical phase

*  analytical phase

*  post-analytical phase

*  future

A.Mosca-UniMI 5 A. Mosca - UniMI 6

* 

* 1. Indicates long-term blood glucose concentration * 2. Fasting not necessary * 3. Sample may be obtained any time of day * 4. Sample stable * 5. Low intra-individual variability * 6. Not altered by acute factors eg, stress,

exercise * 7. Assay standardised across instruments * 8. Concentration predicts the development of

microvascular complications

A. Mosca - UniMI 7

* 

* 1. Cost (?) * 2. May not be available in some areas of world * 3. May be altered by factors other than glucose (e.g.,

change in RBC lifespan, ethnicity) * 4. Some conditions may interfere with measurement * Carbamylated Hb * Labile A1c * Fetal Hb * Hb variants

A. Mosca - UniMi 8

* When HbA1c should not be used for diagnosis

§  Children and adolescents

§  Women until 2 months postpartum

§  People with suspicion of type 1 diabetes (all ages)

§  People with acute diabetes symptoms

§  Prediabetes with acute stress hyperglycemia (f.e. stroke, MI, organ donor)

§  People with drugs (<2 months), which are leading to an acute increase of plasma glucose (glucocorticoids, psychopharmacological drugs)

§  Shortly after pancreas surgery, acute pancreatitis

§  People with HIV infection

§  Endstage renal disease

R. Landgraf (by courtesy)

16/08/18

3

men

women

men

women

Differences man/woman

(Inter99 study) Faerch K et al. Diabetologia 2010;53:858-65

A. Mosca - UniMi 9 A.Mosca-UniMI 10

* HbA1c and age

From 30 y to >70y: Ø  HbA1c increase of 0.5 % • Davidson MB, Schriger DL. Diabetes Res Clin Pract.2010; 87:415–421 • Pani et al.: Diabetes Care 2008; 31:1991–1996 • RaviKumar P et al. Diabetic Med 2011;28:590-594

A. Mosca - UniMi 11Kilpatrick et al, QJMed 1996;89:307

A. Mosca - UniMi 12

16/08/18

4

A. Mosca - UniMi 13

* Races and HbA1c

HbA1c is higher, respect to Caucasian: Ø  Afro-Americans ~ 8 mmol/mol Ø  Latinos ~ 5 Ø  Punjabi Sikhs ~ 4 Ø  Asian ~ 3 • KIRK et al. Diabetes Care 2006; 29:2130–2136 • Likhari T, Gama R. Diabetic Med 2009; 26:1068–1069 • Herman et al. J Clin Endocrinol Metab 2009;94 :1689–1694 • Kamps et al. Diabetes Care 2010;33:1025-1027 • Wolfenbuttel et al. Diabetes Care 2013;36:2931–2936

* Nutrition and HbA1c • Alcohol from abstinence to heavy drinker: HbA1c drop of ~0.4%

The Kaiser Permanente Northern California Diabetes Registry Ahmed et al. J Gen Intern Med 2008; 23:275–82

• The lower the fat consumption and the higher the percentage of unsaturated free fatty acids the lower HbA1c: ~0.2% - 0.3%

Harding et al.: Diabetes Care 2001;24:1911-1916

.

R. Landgraf (by courtesy) A. Mosca - UniMi 14

A.

Mos

ca -

Un

iMI

15

15

Cohen, Blood 2009;112:4284-91

CV~12%

A.Mosca-UniMI 16

16/08/18

5

* Diabetes epidemics and Hemoglobinopathies

A. Mosca - UniMI 17

Diabetes growth 2000 – 2030

Hossain et al, NEJM 2007;356;213-5 Modell et al, Bulletin of the World Health Organization 2008;86:480-7

Hemoglobin disorders incidence

(per 1000 new births)

A.Mosca-UniMI 18

* 

19

of the responders, non-responders and the 7331 patients on the

diabetes register were similar with regard to age, duration of

diabetes and the proportion with Type 1 diabetes (see Table 2).

Twenty-nine per cent of the reference diabetes register

population were known to have a raised urinary albumin

excretion rate (microalbuminuria or dipstick positive

proteinuria), 21% were known to have retinopathy and 31%

had a history of vascular disease. Forty-two per cent were taking

either an angiotensin-converting enzyme inhibitor or

angiotensin II receptor blocker.

The sample was stratified to ensure that we captured people in

all four groups defined by eGFR. This means that some groups

were over-represented, particularly those with the lowest GFR

who are more likely to be anaemic (Table 1). When estimating

the prevalence of anaemia in our population, it was necessary to

adjust the data to take this into account and not overestimate the

total prevalence.

The adjusted prevalence of previously unrecognized anaemia

was 15% of all patients, increasing from 9% in those with an

eGFR ‡ 60 ml ⁄ min per 1.73 m2 to 36% of those with an eGFR

< 60 ml ⁄ min per 1.73 m2 (Table 3 and Fig. 1). A further 17%of

patients reported having a history of anaemia and ⁄ or were taking

medication for anaemia. Eleven patients were taking

erythropoietin and, of these patients, 10 had an eGFR

< 60 ml ⁄ min per 1.73 m2. Of the patients with a history of

anaemia nearly half (44%) were still anaemic. The adjusted total

prevalence of anaemia known and previously unknown was

22%. Five per cent of the study population had a previously

unrecognized anaemia below the recommended treatment

threshold for erythropoietin of 11 g ⁄ dl.

Where previously unrecognized anaemia was detected, its

nature was further characterized by measuring haematinics,

reticulocyte count and erythropoietin levels (Fig. 2). Of those

with a previously unrecognized anaemia, 34% had eryth

ropoietin deficiency, 40% had abnormal haematinics and 26%

had an unexplained anaemia. Seven out of the 10 patients with a

haemoglobin less than 11 g ⁄ dl had erythropoietin deficiency.

Forty-nine per cent of the study respondents were known to be

taking angiotensin-converting enzyme (ACE) inhibitors or

angiotensin II receptor blockers compared with 41% of those

on the diabetes district register. Respondents with anaemia,

either known or previously unknown, were more likely to be

taking ACE inhibitors or angiotensin II receptor blockers (75%).

Table 2 Characteristics of study participants

Responders

n = 234

Non-responders

n = 491

Register

population

n = 7331

Age (years) 61.7 ! 12.7 61.3 ! 15.1 61.8 ! 14.2

Diabetes

duration

(years)

8.8 ! 8.6 8.2 ! 7.9 7.5 ! 7.8

Type 1

diabetes*

10.1 10.8 9.4

Mean ! standard deviation (sd) or %.

*Type 1 diabetes defined as diagnosed under the age of

35 years and treated currently with insulin.

Table 3 Proportion of patients with anaemia

eGFR

(ml ⁄ min

per 1.73 m2)

Unrecognized (%)

Treated

(%)

Total

(%)

WHO

definition*

Haemoglobin

< 11 g ⁄ dl

All subjects† 15 5 7 22

> 90 5 0 0 5

60–90 11 4 5 16

59–30 14 7 9 23

< 30 23 13 23 46

*World Health Organization (WHO) definition of anaemia

haemoglobin < 13 g ⁄ dl for males and < 12 g ⁄ dl for females.

† Prevalence adjusted in relation to reference population dis-

tribution of estimated glomerular filtration rate (eGFR).

(ml/min/1.73 m2) (ml/min/1.73 m2) (ml/min/1.73 m2)

0

5

10

15

20

25

Pro

po

rtio

n w

ith a

nae

mia

()

eGFR > 60 eGFR 30–60 eGFR < 30

WHO definitionHb < 11g/dl

FIGURE 1 Prevalence of previously unrecognized anaemia. WHO

definition*; glycated haemoglobin < 11 g ⁄ dl (?).*World Health

Organization (WHO) definition of anaemia: < 13 g ⁄ dl for males and

< 12 g ⁄ dl for females. eGFR, estimated glomerular filtration rate

AUTHOR: The legend for Figure 1 has been amended - please check .

0

10

20

30

40

50

60

70

EPO Haematinics Unexplained

eGFR > 60eGFR < 60

FIGURE 2 Causes of anaemia in people with diabetes. Estimated

glomerular filtration rate (eGFR) > 60 ml ⁄ min per 1.73 m2; eGFR

< 60 ml ⁄ min per 1.73 m2 (?). EPO, erythropoietin deficiency; Haematinics,

low vitamin B12 or ferritin; Unexplained, anaemia with normal B12, ferritin

MCV anderythropoietin. WorldHealthOrganizationdefinition ofanaemia

< 13 g ⁄ dl for males and < 12 g ⁄ dl for females. eGFR (ml ⁄ min per 1.73 m2)

using Modification of Diet in Renal Disease (MDRD) equation AUTHOR:

The legend for Figure 2 has been amended. Please check.

DIABETICMedicineOriginal article

ª 2010 The Authors.Journal compilation ª 2010 Diabetes UK. Diabetic Medicine, 27, 655–659 657

of the responders, non-responders and the 7331 patients on the

diabetes register were similar with regard to age, duration of

diabetes and the proportion with Type 1 diabetes (see Table 2).

Twenty-nine per cent of the reference diabetes register

population were known to have a raised urinary albumin

excretion rate (microalbuminuria or dipstick positive

proteinuria), 21% were known to have retinopathy and 31%

had a history of vascular disease. Forty-two per cent were taking

either an angiotensin-converting enzyme inhibitor or

angiotensin II receptor blocker.

The sample was stratified to ensure that we captured people in

all four groups defined by eGFR. This means that some groups

were over-represented, particularly those with the lowest GFR

who are more likely to be anaemic (Table 1). When estimating

the prevalence of anaemia in our population, it was necessary to

adjust the data to take this into account and not overestimate the

total prevalence.

The adjusted prevalence of previously unrecognized anaemia

was 15% of all patients, increasing from 9% in those with an

eGFR ‡ 60 ml ⁄ min per 1.73 m2 to 36% of those with an eGFR

< 60 ml ⁄ min per 1.73 m2 (Table 3 and Fig. 1). A further 17%of

patients reported having a history of anaemia and ⁄ or were taking

medication for anaemia. Eleven patients were taking

erythropoietin and, of these patients, 10 had an eGFR

< 60 ml ⁄ min per 1.73 m2. Of the patients with a history of

anaemia nearly half (44%) were still anaemic. The adjusted total

prevalence of anaemia known and previously unknown was

22%. Five per cent of the study population had a previously

unrecognized anaemia below the recommended treatment

threshold for erythropoietin of 11 g ⁄ dl.

Where previously unrecognized anaemia was detected, its

nature was further characterized by measuring haematinics,

reticulocyte count and erythropoietin levels (Fig. 2). Of those

with a previously unrecognized anaemia, 34% had eryth

ropoietin deficiency, 40% had abnormal haematinics and 26%

had an unexplained anaemia. Seven out of the 10 patients with a

haemoglobin less than 11 g ⁄ dl had erythropoietin deficiency.

Forty-nine per cent of the study respondents were known to be

taking angiotensin-converting enzyme (ACE) inhibitors or

angiotensin II receptor blockers compared with 41% of those

on the diabetes district register. Respondents with anaemia,

either known or previously unknown, were more likely to be

taking ACE inhibitors or angiotensin II receptor blockers (75%).

Table 2 Characteristics of study participants

Responders

n = 234

Non-responders

n = 491

Register

population

n = 7331

Age (years) 61.7 ! 12.7 61.3 ! 15.1 61.8 ! 14.2

Diabetes

duration

(years)

8.8 ! 8.6 8.2 ! 7.9 7.5 ! 7.8

Type 1

diabetes*

10.1 10.8 9.4

Mean ! standard deviation (sd) or %.

*Type 1 diabetes defined as diagnosed under the age of

35 years and treated currently with insulin.

Table 3 Proportion of patients with anaemia

eGFR

(ml ⁄ min

per 1.73 m2)

Unrecognized (%)

Treated

(%)

Total

(%)

WHO

definition*

Haemoglobin

< 11 g ⁄ dl

All subjects† 15 5 7 22

> 90 5 0 0 5

60–90 11 4 5 16

59–30 14 7 9 23

< 30 23 13 23 46

*World Health Organization (WHO) definition of anaemia

haemoglobin < 13 g ⁄ dl for males and < 12 g ⁄ dl for females.

† Prevalence adjusted in relation to reference population dis-

tribution of estimated glomerular filtration rate (eGFR).

(ml/min/1.73 m2) (ml/min/1.73 m2) (ml/min/1.73 m2)

0

5

10

15

20

25

Pro

po

rtio

n w

ith a

nae

mia

()

eGFR > 60 eGFR 30–60 eGFR < 30

WHO definitionHb < 11g/dl

FIGURE 1 Prevalence of previously unrecognized anaemia. WHO

definition*; glycated haemoglobin < 11 g ⁄ dl (?).*World Health

Organization (WHO) definition of anaemia: < 13 g ⁄ dl for males and

< 12 g ⁄ dl for females. eGFR, estimated glomerular filtration rate

AUTHOR: The legend for Figure 1 has been amended - please check .

0

10

20

30

40

50

60

70

EPO Haematinics Unexplained

eGFR > 60eGFR < 60

FIGURE 2 Causes of anaemia in people with diabetes. Estimated

glomerular filtration rate (eGFR) > 60 ml ⁄ min per 1.73 m2; eGFR

< 60 ml ⁄ min per 1.73 m2 (?). EPO, erythropoietin deficiency; Haematinics,

low vitamin B12 or ferritin; Unexplained, anaemia with normal B12, ferritin

MCV anderythropoietin. WorldHealthOrganizationdefinition ofanaemia

< 13 g ⁄ dl for males and < 12 g ⁄ dl for females. eGFR (ml ⁄ min per 1.73 m2)

using Modification of Diet in Renal Disease (MDRD) equation AUTHOR:

The legend for Figure 2 has been amended. Please check.

DIABETICMedicineOriginal article

ª 2010 The Authors.Journal compilation ª 2010 Diabetes UK. Diabetic Medicine, 27, 655–659 657

Jones et al. Diabetic Medicine 2010; 27:655-9

A.M

osca-UniMI

n= 7331

A.Mosca-UniMI 20

15/03/18, 08)57Effect of iron supplementation on HbA1c levels in pregnant women with and without anaemia - ScienceDirect

Pagina 1 di 3https://www.sciencedirect.com/science/article/pii/S00098981173…&md5=b8429449ccfc9c30159a5f9aeaa92ffb&dgcid=raven_sd_via_email

Get rights and content

•

•

•

Clinica Chimica ActaVolume 478, March 2018, Pages 57-61

Effect of iron supplementation on HbA1c levels in pregnantwomen with and without anaemiaPaula Breitenbach Renz , Mayana Kieling Hernandez , Joíza Lins Camargo

Show more

https://doi.org/10.1016/j.cca.2017.12.028

Highlights

The effect of iron supplementation (IS) on HbA1c levels in nondiabeticpregnant women was studied.

IS in pregnant women without anaemia or with mild anaemia has noeffect HbA1c levels.

Interpreting HbA1c results in pregnancy during IS and moderate orsevere anaemia still requires caution.

Abstract

Background

a a a, b

Outline

Purchase

Export

Journals Books Register Sign in

16/08/18

6

A.Mosca-UniMI 21

* 

*  references

*  pre-analytical phase

*  analytical phase

*  post-analytical phase

*  future

A.Mosca-UniMI 22

A.Mosca-UniMI 23

Clinical Chemistry

C

o n f

i d e

n t i

a l

Clinical Chemistry

C

o n f

i d e

n t i a

l

A. Mosca - UniMI 24 C.Weykamp(bycourtesy)

16/08/18

7

A.Mosca-UniMI 25

* 

*  references

*  pre-analytical phase

*  analytical phase

*  post-analytical phase

*  future

A.Mosca-UniMI 26

A.Mosca-UniMI 27

42 biochimica clinica, 2011, vol. 35, n. 1

DOCUMENTI DOCUMENTS

PREMESSA

L'emoglobina glicata (HbA1c) è un importantemarcatore biochimico utilizzato nella diagnosi e nelmonitoraggio del diabete, ma la presenza di variantiemoglobiniche può interferire con la sua determinazionerendendo i risultati dell’analisi non attendibili o di difficileinterpretazione (1). Infatti, la presenza di una varianteemoglobinica può determinare sulla misura della HbA1cun'interferenza sia positiva che negativa, che può esseredi tipo strettamente analitico o, più in generale, di tipobiologico/preanalitico. A quest’ultima situazione possonoessere ricondotti i casi in cui sono presenti variantiemoglobiniche che non formano addotti stabili colglucosio o che mostrano cinetiche di glicazione diverserispetto alla HbA o che hanno una vita media eritrocitariaridotta. Si concorda nel ritenere che l'interferenza legataalla presenza di una variante emoglobinica debba esserevalutata per ogni singola variante e per ogni singolometodo utilizzato per la determinazione della HbA1c (2).

VARIANTI EMOGLOBINICHE E INTERFERENZANELLA MISURAZIONE DI HbA1c

Le varianti emoglobiniche più comuni sono HbS,HbC, HbD e HbE. I soggetti omozigoti (HbSS, HbCC,

HbDD) e doppi eterozigoti HbSC presentano un’anemiaemolitica marcata e pertanto, dal momento che laconcentrazione di HbA1c dipende dalla vita eritrocitaria, èopportuno utilizzare parametri differenti per valutare ilcontrollo glicemico in questi pazienti. Inoltre, nei soggettiomozigoti manca totalmente HbA e quindi non si puòformare HbA1c, anche se potrebbe essere presente lacomponente glicata della variante. Nei soggetti portatoridi queste varianti in forma eterozigote, la sopravvivenzaeritrocitaria è normale, anche se in casi isolati pare che isoggetti con eterozigosi HbAC abbiano una vitaeritrocitaria lievemente ridotta (3). La HbA1c può quindiessere utilizzata in questi pazienti se la presenza diqueste varianti non interferisce con il metodo utilizzato.

I metodi attualmente più utilizzati per ladeterminazione della HbA1c possono venire raggruppatisulla base del principio analitico nelle seguenti categorie.

Metodi immunochimici. Vengono utilizzati anticorpiche riconoscono gli ultimi 4-10 amminoacididell’estremità ammino-terminale glicata della catena β.Varianti emoglobiniche che presentano mutazioni cheinteressano questa sequenza amminoacidica (ad es.,HbS e HbC) possono influenzare il risultato della HbA1c(4). Al contrario, la presenza di varianti in cui lamutazione si trova al di fuori della sequenza riconosciutadagli anticorpi (ad es., HbE e HbD) tipicamente non

Refertazione dell’emoglobina glicata in presenza di varianti emoglobiniche

Mariarosa Carta1, Renata Paleari2, Anna Caldini3, Alessandro Terreni3, Andrea Mosca2 per il Gruppo di Studio Intersocietario SIBioC-SIMeL Diabete Mellito1Laboratorio di Chimica Clinica ed Ematologia, ULSS 6, Vicenza2Dipartimento di Scienze e Tecnologie Biomediche, Università degli Studi di Milano3Laboratorio Generale, Azienda Ospedaliero-Universitaria Careggi, Firenze

ABSTRACT

Glycated hemoglobin reporting in presence of hemoglobin variants. Measurement of glycated hemoglobin(HbA1c) is a key-role laboratory test in the management of diabetic patients. Clinicians need reliable and accuratemeasurements, with negligible pre-analytical and post-analytical errors. Among the pre-analytical variables, thepresence of hemoglobin variants is a challenge to the laboratorians: the purpose of this document is to give somepractical advices on how to report HbA1c values in presence of hemoglobin variants. Reporting is quite simple ifmeasurement are carried on by HPLC and other similar analytical methods, when the Hb variant is well separatedand does not interfere with HbA or HbA1c, such as in case of heterozygosis for HbS, HbC, HbD and HbE. In case ofhomozygosis or double heterozygosis for such variants the HbA1c values should not to be reported. Unusual HbA1cvalues, typically <4% or >15% (<20 mmol/mol or >142 mmol/mol), could indicate the presence of hemoglobinvariants. In these cases other approaches to the retrospective evaluation of glycemic control, based on themeasurement of the glycation of different proteins, such as albumin, should be followed.

Corrispondenza a: Andrea Mosca, Dipartimento di Scienze e Tecnologie Biomediche, Università degli Studi, Via Fratelli Cervi 93,20090 Segrate (MI). Tel. 0250330422, Fax 0299987559, E-mail andrea.mosca@unimi.it

Ricevuto: 10.11.2010 Revisionato: 18.11.2010 Accettato: 19.11.2010

Cartaetal,BiochimClin2011;35:42-5A.Mosca-UniMI 28

HbAX

Hbvariantisknown

chekforinterference

Interferenceknown Interferencenotknown

checkifmethodisrobust

yes

HbA1creported

Hbvariantisnotknown

no

stopadvisetouseothertools

HbA1ctoohigh(>140mmol/mol>15%)orHbA1ctoolow(<20mmol/mol<4.0%)

Cartaetal,BiochimClin2011;35:42-5

16/08/18

8

* 

*  references

*  pre-analytical phase

*  analytical phase

*  post-analytical phase

*  future

A.Mosca-UniMI 29 A.Mosca-UniMI 30

A.Mosca-UniMI 31

Contents lists available at ScienceDirect

Clinica Chimica Acta

journal homepage: www.elsevier.com/locate/cca

Average glucose from hemoglobin A1c for altered red blood cell lifetimes:Predictions based on a model for hemoglobin A1c formationRoss Molinaroa,b, Jay H. Hermanc, Douglas F. Sticklec,⁎a Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA, USAb Current affiliation: Siemens Healthineers, Siemens Healthcare Diagnostics, Inc., Tarrytown, NY, USAc Department of Pathology, Jefferson University Hospitals, Philadelphia, PA, USA

A R T I C L E I N F O

Keywords:Hemoglobin A1c

Glycation rateErythrocytesMathematical modelingEstimated average glucoseHemoglobin variants

A B S T R A C T

Background: A model for hemoglobin A1c (HbA1c) formation was used to predict the relationship betweenaverage glucose (AG) and %HbA1c under conditions of altered red blood cell lifetime (RCL).Methods: Using a kinetic mass balance model for formation of HbA1c in red blood cells as a function of age (timein circulation), whole blood %HbA1c vs. glucose was calculated based on the nonlinear distribution of red bloodcells as a function of age across RCL.Results: Model calculations provided a close fit to the standard relationship of estimated average glucose to %HbA1c for normal RCL (r > 0.999). Results for altered RCL were calculated assuming simple time-scale com-pression or expansion of the distribution of red blood cells as a function of RCL. For a given %HbA1c, theoperative average glucose needed to have achieved a given %HbA1c was predicted to be altered by RCL ac-cording to average glucose × RCL = constant.Conclusions: Model calculations estimate the extent to which standard reporting of AG vs. HbA1c underestimatesor overestimates AG under conditions of altered RCL. Conditions of altered RCL may often be operative inpatients with certain hemoglobin variants.

1. Introduction

Serial monitoring of hemoglobin A1c (HbA1c) is a standard of care indiabetes management, as a marker dependent on a patient's averageglucose over the preceding 2–3 month interval [1]. Reporting ofaverage glucose (AG), calculated as a linear function of HbA1c [2], isrecommended by the American Diabetes Association [1]. AG fromHbA1c assumes the general case of normal red blood cell lifetime (RCL,nominally, 120 days). A recognized shortcoming of AG is that it is in-accurate for patients having abnormal RCL, a condition that may beoperative for patients with certain hemoglobin variants [3–7].

A model for HbA1c formation in relation to glucose that is consistentwith data for normal RBC lifetime can provide a basis for predicting therelationship between HbA1c and AG in the case of altered RCL. Here, weuse a simple kinetic mass balance model for HbA1c formation in whichRCL is a parameter. Model parameters were fitted according to con-straints of the standard relationship for AG vs. HbA1c [2]. With thismodel, alterations of RCL in model calculations were used to predict therelationship between AG and HbA1c for altered RCL.

2. Materials and methods

2.1. Model formulation: modeling changes in states of hemoglobin fromnative to glycated forms

The kinetics of HbA1c formation were modeled according to asimple mass balance accounting for three states of a hemoglobin mo-lecule during the course of the lifetime of a red blood cell: A is native(non-glycated) hemoglobin, B is reversibly glycated hemoglobin (aldi-mine, or Schiff base), and C is HbA1c (ketoamine) [8]. Transitions be-tween states were modeled as follows:

A was considered to be in rapid equilibrium with B, dependent on adissociation constant, K = 1040 mmol/l [9], and the glucose con-centration, G, according to:+ = +B (A B) G (G K)

Irreversible transition to C was modeled to be formed at a rateproportional to B according to a first order rate constant, k (= time−1).The overall model is described by following differential equations:+ = − = − + +d(A B) dt k B k G (G K) (A B)

http://dx.doi.org/10.1016/j.cca.2017.09.011Received 8 August 2017; Received in revised form 12 September 2017; Accepted 14 September 2017

⁎ Corresponding author at: Department of Pathology, Jefferson University Hospital, 117 S 11th Street, Pavilion 401, Philadelphia, PA 19107, USA.E-mail address: douglas.stickle@jefferson.edu (D.F. Stickle).

&OLQLFD�&KLPLFD�$FWD���������������²���

$YDLODEOH�RQOLQH����6HSWHPEHU�����������������������(OVHYLHU�%�9��$OO�ULJKWV�UHVHUYHG�

0$5.

Contents lists available at ScienceDirect

Clinica Chimica Acta

journal homepage: www.elsevier.com/locate/cca

Average glucose from hemoglobin A1c for altered red blood cell lifetimes:Predictions based on a model for hemoglobin A1c formationRoss Molinaroa,b, Jay H. Hermanc, Douglas F. Sticklec,⁎a Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA, USAb Current affiliation: Siemens Healthineers, Siemens Healthcare Diagnostics, Inc., Tarrytown, NY, USAc Department of Pathology, Jefferson University Hospitals, Philadelphia, PA, USA

A R T I C L E I N F O

Keywords:Hemoglobin A1c

Glycation rateErythrocytesMathematical modelingEstimated average glucoseHemoglobin variants

A B S T R A C T

Background: A model for hemoglobin A1c (HbA1c) formation was used to predict the relationship betweenaverage glucose (AG) and %HbA1c under conditions of altered red blood cell lifetime (RCL).Methods: Using a kinetic mass balance model for formation of HbA1c in red blood cells as a function of age (timein circulation), whole blood %HbA1c vs. glucose was calculated based on the nonlinear distribution of red bloodcells as a function of age across RCL.Results: Model calculations provided a close fit to the standard relationship of estimated average glucose to %HbA1c for normal RCL (r > 0.999). Results for altered RCL were calculated assuming simple time-scale com-pression or expansion of the distribution of red blood cells as a function of RCL. For a given %HbA1c, theoperative average glucose needed to have achieved a given %HbA1c was predicted to be altered by RCL ac-cording to average glucose × RCL = constant.Conclusions: Model calculations estimate the extent to which standard reporting of AG vs. HbA1c underestimatesor overestimates AG under conditions of altered RCL. Conditions of altered RCL may often be operative inpatients with certain hemoglobin variants.

1. Introduction

Serial monitoring of hemoglobin A1c (HbA1c) is a standard of care indiabetes management, as a marker dependent on a patient's averageglucose over the preceding 2–3 month interval [1]. Reporting ofaverage glucose (AG), calculated as a linear function of HbA1c [2], isrecommended by the American Diabetes Association [1]. AG fromHbA1c assumes the general case of normal red blood cell lifetime (RCL,nominally, 120 days). A recognized shortcoming of AG is that it is in-accurate for patients having abnormal RCL, a condition that may beoperative for patients with certain hemoglobin variants [3–7].

A model for HbA1c formation in relation to glucose that is consistentwith data for normal RBC lifetime can provide a basis for predicting therelationship between HbA1c and AG in the case of altered RCL. Here, weuse a simple kinetic mass balance model for HbA1c formation in whichRCL is a parameter. Model parameters were fitted according to con-straints of the standard relationship for AG vs. HbA1c [2]. With thismodel, alterations of RCL in model calculations were used to predict therelationship between AG and HbA1c for altered RCL.

2. Materials and methods

2.1. Model formulation: modeling changes in states of hemoglobin fromnative to glycated forms

The kinetics of HbA1c formation were modeled according to asimple mass balance accounting for three states of a hemoglobin mo-lecule during the course of the lifetime of a red blood cell: A is native(non-glycated) hemoglobin, B is reversibly glycated hemoglobin (aldi-mine, or Schiff base), and C is HbA1c (ketoamine) [8]. Transitions be-tween states were modeled as follows:

A was considered to be in rapid equilibrium with B, dependent on adissociation constant, K = 1040 mmol/l [9], and the glucose con-centration, G, according to:+ = +B (A B) G (G K)

Irreversible transition to C was modeled to be formed at a rateproportional to B according to a first order rate constant, k (= time−1).The overall model is described by following differential equations:+ = − = − + +d(A B) dt k B k G (G K) (A B)

http://dx.doi.org/10.1016/j.cca.2017.09.011Received 8 August 2017; Received in revised form 12 September 2017; Accepted 14 September 2017

⁎ Corresponding author at: Department of Pathology, Jefferson University Hospital, 117 S 11th Street, Pavilion 401, Philadelphia, PA 19107, USA.E-mail address: douglas.stickle@jefferson.edu (D.F. Stickle).

&OLQLFD�&KLPLFD�$FWD���������������²���

$YDLODEOH�RQOLQH����6HSWHPEHU�����������������������(OVHYLHU�%�9��$OO�ULJKWV�UHVHUYHG�

0$5.

[2] as the basis curve for RCL = 120 days, as would be reported inpractice. Numbers in Table 2 differ slightly from those derived fromFig. 6 because model calculations (non-linear, and having imperfectcorrespondence with the standard linear relationship forRCL = 120 days) were used as the basis curve in Fig. 6.

4. Discussion

We utilized a simple kinetic mass balance model for HbA1c forma-tion to characterize %HbA1c in red blood cells as a function of cell ageand glucose. Whole blood average %HbA1c was predicted from calcu-lation of %HbA1c as a function of cell age combined with a non-uniform

distribution of red blood cell ages. Model predictions for the effects ofRCL on the relationship between %HbA1c and AG yielded a simple re-sult, which was that for a given %HbA1c, a decrease or increase in RCLrelative to normal RCL would require a proportional increase or de-crease in the operative AG, assuming that f(i) scales similarly acrossvarying RCL (per Fig. 1B).

The model is a simplification with respect to physiological detail.For instance: we treated interconversion between A and B componentsas an instantaneous equilibrium, with no kinetic component; we ig-nored the non-zero, pre-circulation value for %HbA1c in reticulocytes;we ignored the possibility of degradation either of hemoglobin or HbA1c

during the red blood cell lifetime; we ignored the possibility of rever-sibility of HbA1c (conversion of C to B or A). Despite these simplifica-tions, model calculation results reasonably represented the standardrelationship of AG vs. %HbA1c for normal RCL = 120 days. Our mainassumption, to enable predictions for the effects of altered RCL on therelationship of AG to %HbA1c, was that the shape of f(i), the distribu-tion of red blood cell ages for any given RCL, was simply a timescale-compressed or timescale-expanded version of the distribution fornormal RCL.

It is generally well known among practitioners that the standardreporting of AG vs. %HbA1c will be an underestimate of AG for con-ditions in which RCL is reduced. However, both pathologists and clin-icians in general are unlikely to be able to estimate the scale on whichthis effect is likely to be operative. The main utility of the modelingexercise is to demonstrate the scale on which this effect is predicted tobe operative, based on the simplest possible physiological model ofHbA1c formation. This may be useful for pathologists in their role asconsultants to other physicians, as a matter of detailed knowledge re-garding the various limitations of HbA1c measurement and interpreta-tion [15]. For instance, results may be useful in tempering interpreta-tion of HbA1c results for patients with certain hemoglobin variants, suchas Hb S and Hb C, for whom RCL is anticipated to be shortened (seeTable 2 in [7]). RCL in S trait has been estimated to be 93 days, eventhough such patients are considered to be hematologically normal [15].Estimates of RCLs for conditions involving hemoglobin variants are,however, regarded with uncertainty [16]; there have been few such

Table 2Projected %HbA1c for RCL = 120 days, compared to measured %HbA1c, as a function of RCL.

1 Measured HbA1c 6 7 8 9 10 11 12 13 14

2 eAG 6.95 8.54 10.13 11.72 13.31 14.90 16.49 18.08 19.67

RCL

140 A oAG 5.96 7.32 8.68 10.05 11.41 12.77 14.13 15.50 16.86B HbA1c(120) 5.4 6.2 7.1 7.9 8.8 9.7 10.5 11.4 12.2

130 A oAG 6.42 7.88 9.35 10.82 12.29 13.75 15.22 16.69 18.16B HbA1c(120) 5.7 6.6 7.5 8.4 9.4 10.3 11.2 12.1 13.0

120 A oAG 6.95 8.54 10.13 11.72 13.31 14.90 16.49 18.08 19.67B HbA1c(120) 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0

110 A oAG 7.58 9.32 11.05 12.79 14.52 16.25 17.99 19.72 21.46B HbA1c(120) 6.4 7.5 8.6 9.7 10.8 11.9 12.9 14.0 15.1

100 A oAG 8.34 10.25 12.16 14.06 15.97 17.88 19.79 21.70 23.60B HbA1c(120) 6.9 8.1 9.3 10.5 11.7 12.9 14.1 15.3 16.5

90 A oAG 9.27 11.39 13.51 15.63 17.75 19.87 21.99 24.11 26.23B HbA1c(120) 7.5 8.8 10.1 11.5 12.8 14.1 15.5 16.8 18.1

80 A oAG 10.43 12.81 15.20 17.58 19.97 22.35 24.74 27.12 29.51B HbA1c(120) 8.2 9.7 11.2 12.7 14.2 15.7 17.2 18.7 20.2

70 A oAG 11.91 14.64 17.37 20.09 22.82 25.54 28.27 30.99 33.72B HbA1c(120) 9.1 10.8 12.6 14.3 16.0 17.7 19.4 21.1 22.8

60 A oAG 13.90 17.08 20.26 23.44 26.62 29.80 32.98 36.16 39.34B HbA1c(120) 10.4 12.4 14.4 16.4 18.4 20.4 22.4 24.4 26.4

1 = measured %HbA1c(NGSP).2 = estimated average glucose (eAG, mmol/L) = 1.59× %HbA1c − 2.59 (standard relationship [2]).RCL = red blood cell lifetime (days). RCL = 120 represents presumed normal RCL, for which oAG = eAG.A = operative average glucose (oAG, mmol/L) = eAG × 120/RCL.B = projected %HbA1c(NGSP) (for RCL = 120 days) = (oAG + 2.59) / 1.59.

Fig. 6. AG vs. HbA1c as predicted according to model calculations with RCL as a para-meter. Compared to RCL = 120, the 95% confidence interval of the standard relationshipfor AG vs. HbA1c (as in Fig. 3) ranges approximately from RCL = 105 to 150 days.

R. Molinaro et al. &OLQLFD�&KLPLFD�$FWD���������������²���

���

A.Mosca-UniMI 32

Contents lists available at ScienceDirect

Clinical Biochemistry

journal homepage: www.elsevier.com/locate/clinbiochem

Use of a model for A1c formation to estimate average glucose operativebetween short-interval measurements of %A1cDouglas F. Sticklea,⁎, Jay H. Hermana, Ross Molinarob,1a Department of Pathology, Kimmel Medical College, Jefferson University, Philadelphia, PA, USAbDepartment of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA, USA

A R T I C L E I N F O

Keywords:Hemoglobin A1cEstimated average glucoseMathematical modelingRed blood cell lifetimeMeasurement intervalsKinetics

A B S T R A C T

Background: Estimated average glucose (AG) is generally reported along with hemoglobin A1c measurementsaccording to a standard calculation. Given a normal red blood cell lifetime of 120 days, serial A1c measurementsat intervals< 120 days are not completely independent. For short interval measurements, a change in AG (ΔAG)necessarily underestimates the change in average glucose operative during the interval (ΔG). We use a model forkinetics of HbA1c to evaluate the theoretical relationship between ΔAG and ΔG for HbA1c measurements madeat intervals between 0 and 120 days.Methods: From any given starting point for A1c, step changes in G were simulated using model calculations todetermine the extent to which A1c could change as a function of the interval of exposure. Values for ΔAG werecompared to the operative ΔG as a function of the interval between A1c measurements.Results: Results of model simulations are a single graph for relationship of ΔAG to ΔG as a function of the intervalbetween A1c measurements. ΔAG for (15, 30, 45, 60, 76, and 90) day intervals underestimated operative ΔG by(73, 51, 34, 21, 11, and 5)%, respectively.Conclusions: Model calculations predict the relationship between changes in estimated average glucose tochanges in operative glucose for serial A1c measurements made at intervals< 120 days. Given that serialmeasurements of A1c made at short intervals are not uncommon in practice, physicians may find this in-formation to be useful.

1. Introduction

As a marker for prevailing glucose, measurement of hemoglobin A1c(as %A1c) up to 4 times per year (every 90 days) is recommended bythe American Diabetes Association standards of care for patients notmeeting treatment goals [1]. Reporting of estimated average glucose(AG) accompanying measurement of %A1c is recommended practice[1], using the AG vs. %A1c correlation of Nathan et al. [2].

In normal subjects, a period of approximately 120 days is needed forthe circulating red blood cell (RBC) population to have been completelyreplaced [3]. For this reason, serial measurements of A1c made at in-tervals that are< 120 days are not fully independent. In practice, serialA1c measurements are often made at intervals that are< 90 days[4–7]. For instance, at Jefferson University Hospital, > 20% of serialmeasurement A1c pairs in 2016 had between-measurement intervalsof< 60 days. Short interval measurements can allow physicians todiscern whether glycemic control is improving or deteriorating. It isdifficult, however, to interpret the extent to which a change in glucose

has been operative in short intervals. Because short interval measure-ments are not fully independent, the difference in reported AG betweenmeasurements must be an underestimate of the change in glucose op-erative during the interval. We present use of a mathematical model forkinetics of A1c formation to characterize maximum changes in %A1cper change in glucose as a function of intervals between measurementsthat are< 120 days, in order to evaluate the theoretical informationcontent of such short interval measurements with respect to the op-erative change in glucose.

2. Methods

2.1. Mathematical model for A1c formation and whole blood A1c fraction

A model for the relationship between AG and whole blood %A1cwas recently presented in the context of a theoretical investigation ofthe effects of variation of RBC lifetime on %A1c [8]. The relationshipwas comprised of two parts: first, a model for the kinetics of formation

https://doi.org/10.1016/j.clinbiochem.2018.02.015Received 29 December 2017; Received in revised form 20 February 2018; Accepted 24 February 2018

⁎ Corresponding author at: Department of Pathology, Jefferson University Hospitals, 117 S 11th St, Pavilion 401, Philadelphia, PA 19107, USA.

1 Current affiliation: Siemens Healthineers, Siemens Healthcare Diagnostics, Inc., Tarrytown, NY, USA.E-mail address: douglas.stickle@jefferson.edu (D.F. Stickle).

&OLQLFDO�%LRFKHPLVWU\�������������²��

$YDLODEOH�RQOLQH����)HEUXDU\�����������������������7KH�&DQDGLDQ�6RFLHW\�RI�&OLQLFDO�&KHPLVWV��3XEOLVKHG�E\�(OVHYLHU�,QF��$OO�ULJKWV�UHVHUYHG�

7

Contents lists available at ScienceDirect

Clinical Biochemistry

journal homepage: www.elsevier.com/locate/clinbiochem

Use of a model for A1c formation to estimate average glucose operativebetween short-interval measurements of %A1cDouglas F. Sticklea,⁎, Jay H. Hermana, Ross Molinarob,1a Department of Pathology, Kimmel Medical College, Jefferson University, Philadelphia, PA, USAbDepartment of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA, USA

A R T I C L E I N F O

Keywords:Hemoglobin A1cEstimated average glucoseMathematical modelingRed blood cell lifetimeMeasurement intervalsKinetics

A B S T R A C T

Background: Estimated average glucose (AG) is generally reported along with hemoglobin A1c measurementsaccording to a standard calculation. Given a normal red blood cell lifetime of 120 days, serial A1c measurementsat intervals< 120 days are not completely independent. For short interval measurements, a change in AG (ΔAG)necessarily underestimates the change in average glucose operative during the interval (ΔG). We use a model forkinetics of HbA1c to evaluate the theoretical relationship between ΔAG and ΔG for HbA1c measurements madeat intervals between 0 and 120 days.Methods: From any given starting point for A1c, step changes in G were simulated using model calculations todetermine the extent to which A1c could change as a function of the interval of exposure. Values for ΔAG werecompared to the operative ΔG as a function of the interval between A1c measurements.Results: Results of model simulations are a single graph for relationship of ΔAG to ΔG as a function of the intervalbetween A1c measurements. ΔAG for (15, 30, 45, 60, 76, and 90) day intervals underestimated operative ΔG by(73, 51, 34, 21, 11, and 5)%, respectively.Conclusions: Model calculations predict the relationship between changes in estimated average glucose tochanges in operative glucose for serial A1c measurements made at intervals< 120 days. Given that serialmeasurements of A1c made at short intervals are not uncommon in practice, physicians may find this in-formation to be useful.

1. Introduction

As a marker for prevailing glucose, measurement of hemoglobin A1c(as %A1c) up to 4 times per year (every 90 days) is recommended bythe American Diabetes Association standards of care for patients notmeeting treatment goals [1]. Reporting of estimated average glucose(AG) accompanying measurement of %A1c is recommended practice[1], using the AG vs. %A1c correlation of Nathan et al. [2].

In normal subjects, a period of approximately 120 days is needed forthe circulating red blood cell (RBC) population to have been completelyreplaced [3]. For this reason, serial measurements of A1c made at in-tervals that are< 120 days are not fully independent. In practice, serialA1c measurements are often made at intervals that are< 90 days[4–7]. For instance, at Jefferson University Hospital, > 20% of serialmeasurement A1c pairs in 2016 had between-measurement intervalsof< 60 days. Short interval measurements can allow physicians todiscern whether glycemic control is improving or deteriorating. It isdifficult, however, to interpret the extent to which a change in glucose

has been operative in short intervals. Because short interval measure-ments are not fully independent, the difference in reported AG betweenmeasurements must be an underestimate of the change in glucose op-erative during the interval. We present use of a mathematical model forkinetics of A1c formation to characterize maximum changes in %A1cper change in glucose as a function of intervals between measurementsthat are< 120 days, in order to evaluate the theoretical informationcontent of such short interval measurements with respect to the op-erative change in glucose.

2. Methods

2.1. Mathematical model for A1c formation and whole blood A1c fraction

A model for the relationship between AG and whole blood %A1cwas recently presented in the context of a theoretical investigation ofthe effects of variation of RBC lifetime on %A1c [8]. The relationshipwas comprised of two parts: first, a model for the kinetics of formation

https://doi.org/10.1016/j.clinbiochem.2018.02.015Received 29 December 2017; Received in revised form 20 February 2018; Accepted 24 February 2018

⁎ Corresponding author at: Department of Pathology, Jefferson University Hospitals, 117 S 11th St, Pavilion 401, Philadelphia, PA 19107, USA.

1 Current affiliation: Siemens Healthineers, Siemens Healthcare Diagnostics, Inc., Tarrytown, NY, USA.E-mail address: douglas.stickle@jefferson.edu (D.F. Stickle).

&OLQLFDO�%LRFKHPLVWU\�������������²��

$YDLODEOH�RQOLQH����)HEUXDU\�����������������������7KH�&DQDGLDQ�6RFLHW\�RI�&OLQLFDO�&KHPLVWV��3XEOLVKHG�E\�(OVHYLHU�,QF��$OO�ULJKWV�UHVHUYHG�

7

Conclusions:ModelcalculationspredicttherelationshipbetweenchangesinestimatedaverageglucosetochangesinoperativeglucoseforserialA1cmeasurementsmadeatintervals<120days.GiventhatserialmeasurementsofA1cmadeatshortintervalsarenotuncommoninpractice,physiciansmayfindthisinformationtobeuseful.

16/08/18

9

A.Mosca-UniMI 33

* Summary*  HbA1c is still the most relevant parameter for monitoring

the glycemic control

*  The Hb variants causing altered red cell lifespan or glycation rate affect results regardless of assay methodology

*  Red cell enzymopathies and red cell membrane disorders may cause also major pre-analytical interference

*  Investigate any result that does not match clinical impression or discordant with other parameters for glucose monitoring.

A. Mosca - UniMI 34

•  Top of the chain: YES (IFCC Network)

•  Middle: probably YES (information not uniform and difficult to be released)

•  Bottom: difficult to draw objective evidences (depending on country and rules);

more efforts needed to achieve standardization overall

A. Mosca - UniMi 35

Standardization of HbA1c: are all the pieces in place?

* •  Always use the best method •  Consensus on interpretation •  Need for more information available

to the laboratory (partioned reference ranges)

• More studies on glycation kinetics •  Education and communication (time

for personalized medicine?)

A. Mosca - UniMI 36

16/08/18

10

A.Mosca-UniMI 37MILANO, ITALYNovember 29th, 2018

AULA MAGNA - LITA SEGRATEVia Fratelli Cervi 93 - Segrate, Milano

Under the auspices of:

GENERAL INFORMATION

CONFERENCE VENUEAULA MAGNA - LITA SEGRATE

Via Fratelli Cervi, 93 - Segrate, Milano

REGISTRATION FEES€ 150,00 + 22% VAT = € 183,00

Special fee for students (including postgraduates)€ 24,60 + 22% VAT = € 30,00

(a document certifying the position is requested)

The registration fee includes:- Lunch buffet as indicated in the programme- Certificate of attendance

Cancellations will be accepted by the organising secretariat strictly within (and no further than) 7 days before the event, entitling the registrant to a 50% reimbursement of the fee paid.Although no reimbursements will be processed further to the cancellation deadline, registration may be turned over to another attendee, provided the name is communicated in advance to the organising secretariat.

To make your registration, please access the following link: http://ems.mzcongressi.com/start/1391/eng

ORGANISING SECRETARIATMZ CONGRESSI srl - Patrizia Sirtori

Via C. Farini, 81 - 20159 - Milano, ItalyPhone: +39 02 66802323 - Fax: +39 02 6686699

patrizia.sirtori@mzcongressi.comwww.mzcongressi.com

12th International Scientific Meeting

STANDARDIZATION IN LABORATORY MEDICINE

AND PATIENT SAFETY

Centro Interdipartimentale per laRiferibilità Metrologica in Medicina di

Laboratorio (CIRME)

Supported by an unconditional grant by

http://users.unimi.it/cirme/home/index.php