1. What is “Diabetes Mellitus”? Without intervention: Complication
Increased plasma glucose conc.
Insulin secretion/resistance index
(disposition index)
Insulin resistance
Without intervention: Complication

2. Evaluation of Insulin Sensitivity / Resistance

3. Control of Plasma Glucose
Pancreas
Adipocytes
Insulin ↑
Urinary glucose
Plasma Glucose
Blood Glucose
200 g/day
Glycogen
Glycogen
Lactic acid
Glycogen
Liver
Glycogen
Muscle
120g/day
Brain
(Food)
Plasma glucose conc. is strictly controlled. Even small increase is out of this control.

4. reduced insulin sensitivity
Insulin resistance
reduced insulin sensitivity
Liver
Muscle
Insulin resistance in the liver, and reduced insulin sensitivity in the muscle are features of type 2 diabetes mellitus (T2DM).

6. Resistance to “Insulin Treatment” Resistance to “Insulin Action”
Insulin Resistance
Resistance to “Insulin Treatment”
Failure to lower plasma glucose conc.
Himsworth HP: Diabetes mellitus: Its differentiation into insulin-sensitive and insensitive types. Lancet i: , 1936.
More than 100 units per day
More than 1.4 units / kg body weight
Flier JS, Kahn CR, Roth J: Receptors, antireceptor antibodies and mechanisms of insulin resistance. N Engl J Med 300:413-9, 1979.
Resistance to “Insulin Action”
Insulin conc. was Higher in diabetic subjects !
Yalow R, Berson S: Immunoassay of endogenous plasma insulin in man. J Clin Invest 39: , 1960.

7. Assessment of Insulin Resistance/Sensitivity
［Indirect methods］
［Direct methods］
Lundbaek K: Intravenous glucose tolerance as a tool in definition and diagnosis of diabetes mellitus. Br Med J 1: , (Insulin: non-steady)
Insulin tolerance test
DeFronzo RA, Tobin J, Andres R: Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 237:E214-E223, 1979.　(insulin & glucose: steady)
Insulin clamp
Turner RC, Holman RR, Matthews D, et al: Insulin deficiency and insulin resistance interaction in diabetes: estimation of their relative contribution by feedback analysis from basal plasma insulin and glucose concentrations. Metabolism 28: , 1979.　(Basal：steady)
HOMA-IR
Bergman RN, Ider YZ, Bowden CR: Quantitative estimation of insulin sensitivity. Am J Physiol 236:E667-E677, 1979.　
(Glucose load: non-steady)
Minimal model

8. Assessment of Insulin Resistance/Sensitivity
［Indirect methods］
［Direct methods］
Lundbaek K: Intravenous glucose tolerance as a tool in definition and diagnosis of diabetes mellitus. Br Med J 1: , (insulin: non-steady)
Insulin tolerance test
DeFronzo RA, Tobin J, Andres R: Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 237:E214-E223, 1979.　(insulin & glucose: steady)
Insulin Clamp
Matsuda M, DeFronzo RA: Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 22: , 1999.　(Basal：steady & Glucose load: non-steady～steady)
Composite Index

9. Direct Evaluation (insulin administration) 1
Therapeutic insulin resistance
Dose of insulin dose of insulin
(Himsworth HP: Lancet i:127, 1936.)
Insulin tolerance test
Change of PG KITT
(Lundbaek K: Brit Med J 1:1507, 1962)
Insulin suppression test
PG SSPG
(Shen S-W et al: J Clin Invest 49:2151, 1970)
Insulin suppression test (modified by somatostatin infusion)
(Harano Y et al: J Clin Endocrinol Metab 45:1124, 1977)
Co-administration of glucose and insulin
IRI and PG during glucose and insulin infusion SSPG
(Johnston C et al: Diabetic Med 7:110, 1990)
次のスライドお願いします。

10. Direct Evaluation (insulin administration) 2 Forearm perfusion
PG, blood flow (PGa-PGv) x Flow
(Zierler K et al: J Clin Invest 40:2111, 1961)
Euglycemic hyperinsulinemic clamp (insulin clamp)
Glucose infusion GIR
(DeFronzo RA et al: Am J Physiol 237:E214, 1979)
Measurement of Ra of glucose by a tracer
Specific activity of glucose, GIR Rd, EGP
(Groop LC et al: J Clin Invest 84:205, 1989)
Pancreatic clamp
Specific activity of glucose, GIR EGP
(Matsuda M et al: Metabolism 51: , 2002)
Analysis of insulin effectiveness
Insulin and glucose infusion ISI-delta, IRI-absolute
(Matsuda M et al: Diabetologia 50:S276, 2007)
次のスライドお願いします。

11. Indirect Evaluation (measurement of plasma insulin concentration) 1
Fasting insulin
Fasting IRI Fasting insulin
(Yalow R et al: J Clin Invest 39:1157, 1960)
Fasting IRI (specific) Fasting insulin
(Olefsky JM et al: Diabetes 22:507, 1973)
HOMA method
Fasting IRI, fasting PG HOMA-IR
(Turner R et al: Diabetologia 28:1086, 1979)
HOMA method updated
Fasting IRI, fasting PG HOMA-%S
(Levy JC et al: Diabetes Care 21:2191, 1998)
QUICKI
Fasting IRI, fasting PG QUICKI
(Katz A et al: J Clin Endocrinol Metab 85:2402)
HOMA-AD (*)
Fasting IRI, fasting PG, adiponectin HOMA-AD
(Matsuhisa M et al: Diabetes Res Clin Pract, 77:151, 2007)
*: Regressed algorithm
For sensitivity; 1/FI
次のスライドお願いします。

12. Indirect Evaluation (measurement of plasma insulin concentration) 2
CIGMA
IRI and PG during glucose infusion CIGMA
(Hosker JP et al: Diabetologia 28:401, 1985)
OGTT product
IRI and PG during oral glucose load average PG x average IRI
(Levine R et al: N Engl J Med 283:237, 1970) SI
IRI and PG during oral glucose load SI
(Cederholm J et al: Diabetes Res Clin Pract 10:167, 1990)
Composite index or Matsuda index
IRI and PG during oral glucose load ISI(Comp)
(Matsuda M et al: Diabetes Care 22:1462, 1999)
Si calculated from fasting and post oral glucose load
IRI and PG during oral glucose load Sib, Si2h, SiM
(Avignon A et al: Int J Obes Relat Metab Disord. 23:512, 1999)
ISI (*)
BMI, IRI at 120 min, PG at 90 min ISI
(Stumvoll M et al: Diabetes Care 23:295, 2000)
*: Regressed algorithm
次のスライドお願いします。

13. Indirect Evaluation (measurement of plasma insulin concentration) 3
OGIS
IRI and PG after oral glucose load OGIS
(Mari A et al: Diabetes Care 24:539, 2001)
SI (oral) : minimal model index
IRI and PG during meal load SI (oral)
(Caumo A et al: J Clin Endocrinol Metab 85:4396, 2000)
IRI and PG during oral glucose load SI (oral)
(Brenda E et al: Diabetes 50:150, 2001)
SIis OGTT
IRI and PG after oral glucose load SIis
(Bastard JP et al.: Diabetes Metab 33:261, 2007)
Minimal model (FSIVGTT)
PG and IRI after iv glucose infusion SI
(Bergman RN et al: Am J Physiol 236:E667, 1979)
Minimal model (dynamic analysis)
PG and IRI after iv glucose infusion SID
(Pillonetto G et al: IEEE Trans Biomed Eng 53:369, 2006)
*: Regressed algorithm
次のスライドお願いします。

14. Induction of HOMA-IR (1)
Insulin sensitivity
positive constant
plasma glucose, insulin conc.
the fractional disappearance rate of glucose
(insulin action)
the volume of distribution of glucose
the glucose input rate
(Radziuk J: J Clin Endocrinol Metab 85: , 2000)

15. Induction of HOMA-IR (2)
Steady state:
Insulin sensitivity　（steady state）
(Radziuk J: J Clin Endocrinol Metab 85: , 2000)

16. After glucose administration
MCR (metabolic clearance rate)
Dose of glucose
AUC of PG conc.
(non- steady state) = PG
Glucose Dose
mean
MCR
~180min

17. After glucose administration
Insulin Sensitivity during OGTT
MCR of glucose
Average Insulin conc.
Dose of glucose
　PG ×　Insulin
can be estimated by =

18. Induction of Composite Index ISI(comp) Inverse of Geometric Mean
次のスライドお願いします。
Inverse of Geometric Mean

19. ∫ ∫ 10,000 (FPG X FPI)X(G X I) ISI(comp)= G= g(t) dt I= i(t) dt
Insulin sensitivity indices obtained from oral glucose tolerance testing
10,000
ISI(comp)=
(FPG X FPI)X(G X I)
次のスライドお願いします。
240 1 ∫ G=
g(t) dt
mean
240 1
240 ∫ I=
i(t) dt
240
240
Response up to 240 minutes or time that is necessary to reach basal state.
Matsuda M, DeFronzo RA.: Diabetes Care 22(9): , 1999.

20. ∫ ∫ 10,000 (FPG X FPI)X(G X I) ISI(comp)= G= g(t) dt I= i(t) dt
Insulin sensitivity indices obtained from oral glucose tolerance testing
10,000
ISI(comp)=
(FPG X FPI)X(G X I)
次のスライドお願いします。
120 1 ∫ G=
g(t) dt
mean
120 1
120 ∫ I=
i(t) dt
120
120
Usually we have OGTTs with 120 minutes. And the initial validation was done from the data up to 120 min.
Matsuda M, DeFronzo RA.: Diabetes Care 22(9): , 1999.

21. ISI(comp) (Composite Index, Matsuda’s Index)
Mean (range, ±SD) in healthy young persons
New Haven, CT (n=37) ( , ±1.9)
San Antonio, TX (n=62) ( , ±2.6)
Correlation with clamp: r ≧ 0.73
Matsuda M, DeFronzo RA: Insulin sensitivity indices obtained from oral glucose tolerance testing. Comparison with the euglycemic insulin clamp. Diabetes Care 22: , 1999.

23. Cut off value for Insulin Resistance
2,321 (2,138 nondiabetic) euglycemic insulin clamp studies
European Group for the Study of Insulin Resistance (EGIR) project (n = 1,436)
Pima Indian Study (n = 597)
San Antonio (n = 288, of whom 99 were Mexican American)
The EGIR studies were performed on Caucasians from 17 European sites (Athens, Greece; Baden, Heidelberg, Kreisha, and Munich,Germany; Belgrade, Serbia; Geneva, Switzerland; Goteborg, Sweden; Helsinki and Kuopio, Finland; Odense, Denmark; and Naples, Padova, Pisa, Rome,Torino, and Verona, Italy).
Mixture model for whole-body glucose disposal measured by the euglycemic insulin clamp technique. Dashed line, histogram including diabetic subjects; solid line, histogram excluding diabetic subjects; dotted line, normal mixture density estimate.
Cut off Value
Rd: <28 mmol/min per kg LBM
　　　　(5.0 mg/min per kg LBM)
　　　　(200 mg/min per m2)
ISI(comp): < ~ 3.1
HOMA-IR: > ~ 2.8
Diagnosis of insulin resistance
(if any of the following conditions are met)
BMI >28.9 kg/m2
HOMA-IR>4.65
BMI >27.5 kg/m2
and HOMA-IR >3.60
(Sensitivity 84.9%, Specificity 78.7%)
Insulin Sensitivity
Diabetes 54:333–339, 2005

24. Cut off Value Calculation
from the data set used to introduce ISI(comp)
Rd<200mg/m2 per min as the gold standard
Sensitivity Specificity
BMI≧ % 65%
BMI≧ % 80%
(Rd vs BMI: r=-0.45)
HOMA-IR≧ % 69%
HOMA-IR≧2.80* 80% 77%
(Rd vs HOMA-IR: r=-0.44)
ISI(comp)＜3.1 76% 71%
ISI(comp)＜2.6* 71% 84%
(Rd vs ISI(comp): r=0.73)
Model 91% (85%¶) 62% (79%¶)
BMI >28.9 kg/m2 or HOMA-IR>4.65 or (BMI >27.5 kg/m2 and HOMA-IR >3.6)
*:best value by ROC analysis
¶: shown in Diabetes 54:333–339, 2005
(Data set used in Diabetes Care 22: 1462, 1999)
ISI(comp)=2.5 was used as a cut-off in Kerman WN, et al: Pioglitazone improves insulin sensitivity among non diabetic patients with a recent transient ischemic attach or ischemic stroke. Stroke 34:1431-6, 2003

26. Homeostatic Model Assessment
HOMA
Homeostatic Model Assessment

27. Anyone Should Appreciate the Danger of Interpreting Ratios
Is it possible that inulin resistance can be assessed by the ratio of fasting plasma insulin to fasting plasma glucose concentration? A:
PG 90mg/dl
IRI 5mU/ml
Ratio 90/5=18
Product 90x5=450 B:
IRI 10mU/ml
Ratio 90/10=9
Product 90x10=900 C:
PG 120mg/dl
Ratio 120/10=12
Product 120x10=1200
次のスライドお願いします。
The order of insulin resistance has to be C, B, A!
Insulin Resistance can be assessed by
the product of FPG and fasting insulin conc.

28. Description of the Original HOMA
PERIPHERAL
TISSUE
Description of the Original HOMA
BRAIN
次のスライドお願いします。
Metabolism 28: , 1979

29. Homeostasis model analysis (HOMA)
First description in 1979 Turner et al.
Turner R, Holman RR, Matthews D, Hockaday TR, Peto J : Insulin deficiency and insulin resistance interaction in diabetes : estimation of relative contribution by feedback analysis from basal plasma insulin and glucose concentrations.
(Metabolism 28: , 1979.)
HOMA INDEX(
Insulin resistance
)=rl=rp
During basal steady state: 0=
HGP & Splanchnic ( f
PG,fIRI,rl) －
Brain
(fPG) －
Muscle
(fPG,fIRI, rp)
次のスライドお願いします。
This is converted to the below formula. The unit of PG is mmol/min.
fIRI
fIRI
0.4
0.4
1.2
1.2 1 1 1 1 0= 0=
3-1.86
3-1.86 × ×
log
log
-1.5
-1.5 × ×
log(fPG)
log(fPG) - - - - × × × × rl rl
0.1
0.1 14 14
0.4
0.4 6 6 1+ 1+ 1+ 1+ 1+ 1+
fPG
fPG
fIRI
fIRI
fPG
fPG +2 +2 rp rp
Where
fIRI : fasting insulin conc. [mU/L], fPG : fasting PG [mmol/L]
rl : liver insulin resistance, rp : peripheral (or muscle) insulin resistance
Assuming that rl=rp(=R), the above formula can be solved for R after substituting fPG and fIRI for actual measured values.
NOTE: the function above is in steady state.

30. Calculation of HOMA-IR (example)
fIRI
fIRI
0.4
0.4
1.2
1.2 1 1 1 1 0= 0=
3-1.86
3-1.86 × ×
log
log
-1.5
-1.5 × ×
log(fPG)
log(fPG) - - - - × × × × rl rl
0.1
0.1 14 14
0.4
0.4 6 6 1+ 1+ 1+ 1+ 1+ 1+
fPG
fPG
fIRI
fIRI
fPG
fPG
rl=rp(=R)，fIRI=8 [mU/ml] , fPG=6 [mmol/L] +2 +2 rp rp
Then, R 8 1+
0.4
0.1 6 14 +2
1.2 1 0=
3-1.86 ×
log
-1.5
log(6) -
次のスライドお願いします。
R=2.1　　(when fIRI=8 mU/ml, fPG=108mg/dl(=6mmol/L))
Thus,
Note: this calculation was estimated from the figures found in the original Metabolism paper published in It may possible that Dr. Turner used a different formuola.
It was possible to solve this equation by a large frame computer in Now it is easy to solve this equation by an EXCEL file.
M. Matsuda: index of insulin secretion and insulin resistance, Internal Medicine [Japanese] 105:39-44, 2010.

31. Induction of Simplified HOMA formula
Reduced formula
(1985 by Matthews D et al.)
HOMA-IR=
次のスライドお願いします。
HOMA-b%=
Diabetologia 28: , 1985

32. QUICKI
Kats A, Nambi SS, Mather K et al: Quantitative Insulin-Sensitivity Check Index (QUICKI): a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 85: , 2000.

33. INSULIN DOSE RESPONSE CURVEＳｓ
INSULIN DOSE RESPONSE CURVE
Difference found in HOMA-IR or QUICKI
Total Glucose Disposal
(mg/min・kg)
Production (mg/min・kg)
Hepatic Glucose
Rd(t)
Ra(t)
Basal glucose metabolism ： ~2.2 mg/kg per min
Plasma insulin concentration was conc. in portal vein for hepatic glucose production.

34. HOMA-IR vs Insulin Clamp
Differences in Insulin Clamp
緑のグラフは正常対象者の糖利用のインスリン用量依存曲線であります。
さて，筋肉はインスリンに抵抗性があるのかは，インスリン濃度を変化させて
その用量依存曲線が右下の黄色いグラフの方向にゆけば抵抗性があると言ってよろしいと思います。
実際に，肥満者や糖尿病患者では抵抗性があるわけです。
次のスライドお願いします。
Differences in HOMA-IR or QUICKI
Insulin (mU/ml)