ArticlePDF Available

Prospective Relation of C-Reactive Protein With Type 2 Diabetes: Response to Han et al

June 2003
Diabetes Care 26(5):1656-7; author reply 1657-8

June 2003
26(5):1656-7; author reply 1657-8

DOI:10.2337/diacare.26.5.1656

Source
PubMed

Authors:

Marieke Snijder

Academisch Medisch Centrum Universiteit van Amsterdam

Show all 9 authorsHide

No caption available

…

Figures - uploaded by Jaap Seidell

Content may be subject to copyright.

Content uploaded by Jaap Seidell

Content may be subject to copyright.

OBSERVATIONS

Analysis of the

Agreement Between

the World Health

Organization

Criteria and the

National Cholesterol

Education Program-

III Deﬁnition of the

Metabolic Syndrome

Results from a population-based

survey

ur objective was to evaluate the di-

agnostic proﬁciency of the World

Health Organization (WHO) and

the National Cholesterol Education Pro-

gram (NCEP)-III deﬁnitions (1,2) for the

metabolic syndrome in a Mexican nation-

wide, population-based survey. Details of

the sampling procedures have been pre-

viously described (3). The population was

composed of 2,158 men and women aged

20–69 years sampled after a 9- to 12-h

fasting period. For the WHO criteria, in-

sulin resistance was diagnosed if a non-

diabetic case had fasting insulin concen-

trations ⱖ126 pmol/l (21 ␮U/ml) (⬎75th

percentile in Mexican adults). The age-

adjusted prevalence was 13.61% for the

WHO criteria (n⫽268) and 26.6% for

the NCEP-III deﬁnition (n⫽574). After

excluding patients with diabetes, the prev-

alence was 9.2 and 21.4%, respectively.

The agreement between the deﬁnitions

was assessed in 1,969 subjects; 189 cases

were eliminated due to the lack of a urine

sample.

The number of abnormal cases was

lower using the WHO criteria. Only 237

of the 545 subjects (43.4%) who fulﬁlled

the NCEP criteria were diagnosed as af-

fected using the WHO deﬁnition. Just 16

of 253 cases (6.3%) detected by the WHO

deﬁnition did not fulﬁll the NCEP deﬁni-

tion. The agreement between the criteria

was moderate (␬⫽0.507). On the other

hand, the subjects diagnosed using the

WHO recommendations had a worse

proﬁle than the cases detected by the

NCEP-III deﬁnition only—they had a

higher BMI and higher non-HDL choles-

terol, triglyceride, and glucose concentra-

tions. The demonstration of insulin

resistance among the nondiabetic popu-

lation caused the lack of agreement in 202

of the 242 cases that fulﬁlled the NCEP

deﬁnition but failed the WHO criteria.

Other reasons for disparity were the

higher thresholds used by the WHO cri-

teria; these differences explained the lack

of agreement in 66 of the 152 cases with

diabetes.

In conclusion, the prevalence of the

metabolic syndrome is inﬂuenced by the

selection of the diagnostic criteria. The

WHO criteria identiﬁed a lower number

of cases than the NCEP-III deﬁnition.

These differences were explained mainly

by the inclusion of abnormally high insu-

lin concentrations as a diagnostic crite-

rion. However, the presence of insulin

resistance may help to identify patients

more severely affected (4).

CARLOS A. AGUILAR-SALINAS,

ROSALBA ROJAS,

PHD

FRANCISCO J. GO

´MEZ-PE

´REZ,

VICTORIA VALLES,

JUAN MANUEL Rı´OS-TORRES,

AURORA FRANCO,

PHD

GUSTAVO OLAIZ,

PHD

JUAN A. RULL,

JAIME SEPULVEDA,

PHD

From the

Departamento de Endocrinologı´a y Me-

tabolismo del Instituto Nacional de Ciencias Me´di-

cas y Nutricio´n “Salvador Zubira´n,” Mexico City,

Me´xico; and the

Instituto Nacional de Salud Pu´-

blica, Cuernavaca, Morelos, Me´xico.

Address correspondence to Carlos Alberto Agui-

lar-Salinas, MD, Vasco de Quiroga 15, Mexico City

14000, Me´ xico. E-mail: caguilarsalinas@yahoo.com.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Expert Panel on Detection, Evaluation,

and Treatment of High Blood Cholesterol

in Adults: Executive Summary of The

Third Report of The National Cholesterol

Education Program (NCEP) Expert Panel

on Detection, Evaluation, and Treatment

of High Blood Cholesterol In Adults

(Adult Treatment Panel III). JAMA 285:

2486–2497, 2001

2. Alberti FGMM, Zimmet PZ: Deﬁnition,

diagnosis and classiﬁcation of diabetes

mellitus and its complications. Part 1: di-

agnosis and classiﬁcation of diabetes mel-

litus provisional report of a WHO con-

sultation. Diabet Med 15:539–553, 1998

3. Aguilar-Salinas CA, Rojas R, Go´mez-Pe´ rez

FJ, Garcı´a E, Valles V, Rı´os-Torres JM,

Franco A, Olaiz G, Sepu´lveda J, Rull JA:

Prevalence and characteristics of early-

onset type 2 diabetes in Mexico. Am J Med

113:569–574, 2002

4. Hanson R, Imperatore G, Bennett PH,

Knowler W: Components of the “meta-

bolic syndrome” and incidence of type 2

diabetes. Diabetes 51:3120–3127, 2002

Complete Blockade

of the Renin-

Angiotensin System

in Patients With

Advanced Diabetic

Nephropathy

e read with interest the article by

Rossing et al. (1) in the January

2002 issue of Diabetes Care.We

had previously assessed the effect of com-

plete blockade of the renin-angiotensin

system (RAS) in patients with advanced

diabetic nephropathy (type 2 diabetes)

and severe proteinuria (2). To do so, we

studied 10 patients on prior treatment

with ACE inhibitors at recommended

doses, to which 50 mg/day of Losartan

was added.

After 3 months of the combined ther-

apy, urinary protein excretion decreased

from a mean of 6.9 (95% CI 4.3–9.6) to

5.8 g/24 h (3–8.6) (P⫽0.025) and, with

the exception of two patients, was re-

duced in all cases. The proteinuria/

urinary creatinine ratio also decreased

from 7.6 (4.2–11.05) to 6.0 g/g (3.2–8.7)

(P⫽0.02). In one patient in which pro-

teinuria did not vary, the proteinuria/

creatinine ratio also decreased, reﬂecting

a possible error in the quantiﬁcation of

the patient’s proteinuria instead of a lack

of response to the treatment. The glomer-

ular ﬁltration rate (GFR), calculated using

the mean of creatinine and urea clearance,

a measurement that avoids the overesti-

mation that creatinine clearance produces

in the GFR (3), decreased slightly from

21.5 ⫾8.5 to 19.9 ⫾7.8 ml/min (P⫽

0.158). An interesting decrease in total

cholesterol from 6.21 ⫾1.03 to 5.33 ⫾

1.27 mmol/l (P⫽0.036) was observed.

Total proteins, serum albumin, protein

intake, and mean arterial pressure (MAP)

were not signiﬁcantly modiﬁed. Only one

patient had hyperkaliemia (from 5.1 to

7.2 mmol/l). No secondary effects were

seen in the other patients. At 12 months of

treatment, proteinuria decreased to 3.4

LETTERS

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1635

g/24 h and the proteinuria/urinary creat-

inine ratio to 3.6. GFR decreased by 0.83

ml 䡠min

⫺1

䡠month

⫺1

in the ﬁrst 6

months and only fell by 0.18 ml 䡠min

⫺1

䡠

month

⫺1

during the last 6 months. Two

patients received the combined treatment

for ⬎42 months. During the ﬁrst 18

months, proteinuria decreased by 74 and

67% in each case; GFR fell by 0.5 and 1 ml

䡠min

⫺1

䡠month

⫺1

, respectively; and MAP

did not vary in the ﬁrst case and decreased

from 123 to 101 mmHg in the second.

Over the following 24 months, protein-

uria was ⬍1 g/24 h in both cases and GFR

increased in one patient by 2 ml/min and

did not vary in the other patient. In this

period, MAP and protein intake remained

unvaried.

The data of Rossing et al. conﬁrm that

in the short term, complete blockade of

the RAS system produces an antiprotein-

uric effect and affords a greater renopro-

tective effect since proteinuria is an

important risk factor for kidney disease

progression (4). Although our data can

only be considered as observational ﬁnd-

ings, they do allow us to intuit that this

renoprotective effect, due to the dual

blockade of the RAS, will be increased

when treatment is prolonged to medium

or long term. Moreover, the observed de-

crease in total cholesterol levels would

contribute to reducing both the progres-

sion of the nephropathy, since elevated

serum cholesterol levels act as an inde-

pendent promoter of progression in dia-

betic nephropathy (5), and the high

cardiovascular risk seen in this type of

patient.

JAVIER DEIRA,

MD, PHD

HUGO DIAZ,

JESU

´SGRANDE,

MD, PHD

From the Division of Nephrology, Internal Medi-

cine, Hosptial Virgen de la Concha, Zamora, Spain.

Address correspondence to Dr. Javier Deira, Di-

vision of Nephrology, Internal Medicine, Hospital

Virgen de la Concha, Avenida de Requejo 35, 49022

Zamora, Spain. E-mail: jdeira@saludalia.com.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Rossing K, Christensen PK, Jensen BR,

Parving H-H: Dual blockade of the renin-

angiotensin system in diabetic nephropa-

thy. Diabetes Care 25:95–100, 2002

2. Deira J, Dı´az H, Gonza´lez A, Grande J: Is

there a synergic antiproteinuric and reno-

protective efect between angiotensin-con-

verting-enzyme inhibitors and angiotensin

type1 receptor antagonists in patients

with advanced diabetic nephropathy (Ab-

stract). J Hypertens 19 (Suppl. 2):S137,

2001

3. Kasiske BL, Keane WF: Laboratory assess-

ment of renal disease: clearance, urinaly-

sis, and renal biopsy. In The Kidney. 5th

ed. Brener BM, Rector FC, Eds. Philadel-

phia, WB Saunders, 1996, p. 1140

4. Remuzzi G: Nephropathic nature of pro-

teinuria. Curr Opin Nephrol Hypertens

8:655–663, 1999

5. Hovind P, Rossing P, Tarnow L, Smidt

UM, Parving HH: Progression of diabetic

nephropathy. Kidney Int 59:702–709, 2001

Effect of 2-Week

Treatment With

Pirenzepine on

Fasting and

Postprandial Glucose

Concentrations in

Individuals With

Type 2 Diabetes

cute cholinergic muscarinic block-

ade with pirenzepine (PIR) induces

dose-related reductions in plasma

glucose (PG) and plasma insulin (PI) after

mixed meals in normal individuals and in

type 2 diabetic and polycystic ovary syn-

drome patients (1–2). Altogether, 24 pa-

tients with type 2 diabetes (12 male and

12 female) with a fasting plasma glucose

ⱖ10 mmol/l were studied for the effects

of PIR (50 mg b.i.d.) given for 2 weeks on

fasting and postprandial plasma glucose

and insulin concentrations. Patients on

ACE inhibitors or thiazide diuretics were

excluded.

Oral hypoglycemics were stopped at

least 2 weeks before treating the patients

with either PIR or placebo in a random-

ized, cross-over, double-blind fashion

with a 2-week washout period between

the two treatment arms. At the beginning

(day 1) and end (day 14) of each treat-

ment period, blood was sampled over

12 h, during which time the patients re-

ceived three standard meals. The sched-

uling for blood sampling, meals, and PIR

administration is illustrated in Fig. 1. The

primary outcome measure for the study

was the difference between PIR- and pla-

cebo-treated fasting plasma glucose on

day 14 of the study. Statistical analysis

was performed using the paired Student’s

ttest.

There was a signiﬁcant reduction in

the fasting PG levels on day 14 (PIR) com-

pared with day 14 (placebo) (mean reduc-

tion ⫾SD, 1.2 ⫾1.6 mmol/l, 95% CI

0.5–1.9, P⫽0.002 by paired Student’st

test) (Fig. 1). There were trends for reduc-

tion in the glucose area under the curve

and highest postprandial peaks for day 14

(PIR) compared with day 14 (placebo),

but the results did not reach statistical sig-

niﬁcance. There was no signiﬁcant differ-

ence in fasting plasma insulin levels.

The mechanism through which PIR

produces its effect on PG is unknown.

Possible hypotheses are the obliteration of

an increased hypothalamic cholinergic

tone responsible for increased 24-h growth

hormone (GH) secretion, particularly

suppression of nocturnal GH secretion,

which we and others (3,4) have previ-

ously demonstrated. Another possibility

is the effect of PIR on gastrointestinal hor-

mones through its anticholinergic effect,

which may affect PG levels indirectly (5).

In conclusion, PIR caused a statisti-

cally signiﬁcant reduction in fasting PG

levels compared with placebo after 2

weeks of therapy, but there was no corre-

sponding increase in insulin levels. PIR

may be an effective drug in the treatment

of type 2 diabetes.

Acknowledgments—This study was sup-

ported by a grant from Boehringer Ingelheim.

We are indebted to Drs. Mary Lewis and

Mark Lewis for their assistance with the insu-

lin assay and Liz Gardner for her assistance

with statistical analysis.

BASIL G. ISSA,

NICHOLA DAVIES,

RGN

KERENZA HOOD,

PHD

LAKDASA D.K.E. PREMAWARDHANA,

JOHN R. PETERS,

MAURICE F. SCANLON,

From the

Department of Endocrinology, University

Hospital of Wales, Heath Park, Cardiff, U.K.; the

Department of Statistics, University Hospital of

Wales, Heath Park, Cardiff, U.K.; and the

Depart-

ment of Medicine, Caerphilly Miners District Hos-

pital, Caerphilly, Mid Glamorgan, U.K.

Address correspondence to Dr. Basil G. Issa, MD,

FRCP, MRCP(I), Diabetes Centre, North Manches-

ter General Hospital, Delaunays Road, Crumpsall,

Manchester M8 5RB U.K. E-mail: basil-issa@mail.

nmanhc-tr.nwest.nhs.uk.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Bevan JS, Ara J, Page MD, Scanlon MF,

Peters JR: Cholinergic blockade with

pirenzepine induces dose-related reduc-

Letters

1636 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

tion in glucose and insulin responses to a

mixed meal in normal subjects and non-

insulin dependent diabetics. Clin Endocri-

nol 35:85–91, 1991

2. Premawardhana LDKE, Ismail IS, Riad-

Fahmy D, Miell JP, Peters JR, Scanlon MF:

Acute cholinergic blockade with low dose

pirenzepine reduces the insulin and glucose

responses to a mixed meal in obese women

with the polycystic ovary syndrome. Clin

Endocrinol 40:617–621, 1994

3. Page MD, Bevan JS, Dieguez C, Peters JR,

Scanlon MF: Cholinergic blockade with

pirenzepine improves carbohydrate toler-

ance and abolishes the GH response to

meals in normal subjects. Clin Endocrinol

30:519–524, 1989

4. Pietschmann P, Schernthaner G: The ef-

fect of pirenzepine on growth hormone

and blood glucose levels in type I diabetes

mellitus: a controlled study of patients on

basal bolus insulin treatment. Acta Endo-

crinologica (Copenh) 117:315–319, 1988

5. El-Sabbagh HN, Bloom SR, Adrian TE,

Prinz RA, Baron JH, Welbourne RB: The

effect of pirenzepine on meal-stimulated

gastrointestinal hormones. Scandinavian J

Gastroenterol 66:57–61, 1980

Improved Insulin

Sensitivity and

Metabolic Control in

Type 2 Diabetes

Does Not Inﬂuence

Plasma Homocysteine

ype 2 diabetes is characterized by

both insulin resistance and an in-

creased cardiovascular mortality

(1). Elevated plasma total homocysteine

levels are an independent risk factor for

cardiovascular disease (2). Also, in pa-

tients with type 2 diabetes, elevated

plasma total homocysteine levels are asso-

ciated with an increased risk of cardiovas-

cular disease and an increased mortality

(3,4).

It is well known that insulin resis-

tance is aggravated by chronic hypergly-

cemia. This hyperglycemia-induced

insulin resistance is reversed by treat-

ment, resulting in a prolonged period of

euglycemia (5).

Although in animal studies evidence

was found that hyperinsulinemia and/or

insulin resistance increased total homo-

cysteine levels (6), in humans the relation

between insulin resistance and total ho-

mocysteine levels is unclear, in both

healthy subjects (7–12) and patients with

type 2 diabetes (13–16). In patients with

type 2 diabetes, a positive association be-

tween insulin resistance and total homo-

cysteine levels was reported (13), but this

ﬁnding was not reproduced (14). Fur-

thermore, in patients with type 2 diabe-

tes, a positive association between the

degree of metabolic control and total ho-

mocysteine levels was reported (15,16).

We hypothesized that if insulin resis-

tance is associated with elevated homo-

cysteine levels, thereby providing another

explanation for the increased cardiovas-

cular disease in type 2 diabetes, then ho-

mocysteine levels will decrease after

amelioration of hyperglycemia-induced

insulin resistance and associated meta-

bolic abnormalities.

Here, we report the results of an in-

tervention study (17). Eight obese type 2

diabetic patients (aged 53 ⫾13 years,

HbA

12.0 ⫾1.7%, BMI 38 ⫾5.8 kg/

) with severe insulin resistance (subcu-

taneous insulin dose, four-dose insulin

regimen 1.92 ⫾0.66 units 䡠kg

⫺1

䡠day

⫺1

)

were studied before and after a period of

28 ⫾5 days of intravenous insulin treat-

ment. Intravenous insulin treatment re-

sulted in euglycemia. Before and after

intervention, insulin sensitivity was as-

sessed by a hyperinsulinemic-euglycemic

clamp. Fasting plasma total homocys-

teine, HbA

, and lipid concentrations

were measured on the days of the clamps.

Also, 24-h urine creatinine excretion was

investigated before and after intervention.

Plasma total homocysteine levels

were measured with a high-performance

liquid chromatography method accord-

ing to Fiskerstrand (18,19) with some

modiﬁcations. Because all variables were

normally distributed, statistical analyses

of differences were performed by paired

Student’sttest. P⬍0.05 was considered

Figure 1—Day curve of plasma glucose (A) and insulin (B) concentrations after 14 days of PIR

(50 mg b.i.d.) (F) and placebo (E). The vertical continuous lines represent the time when piren-

zepine was administered and the interrupted lines represent the times at which breakfast, lunch,

and dinner were given.

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1637

signiﬁcant. Results are given as the

mean ⫾SD.

After a 4-week period of intravenous

insulin treatment resulting in an euglyce-

mic period of 17 ⫾4 days, insulin sensi-

tivity improved since body glucose

uptake, measured by clamp, increased

(12.7 ⫾5.7 before vs. 22.4 ⫾8.8 ␮mol 䡠

⫺1

䡠min

⫺1

after euglycemia, P⬍

0.0005) and the intravenous insulin dose

required for achieving and maintaining

euglycemia decreased (1.7 ⫾0.9 before

vs. 1.1 ⫾0.6 units 䡠kg

⫺1

䡠day

⫺1

after

euglycemia, P⬍0.005). Metabolic con-

trol improved; HbA

decreased substan-

tially (12.0 ⫾1.7 before vs. 8.9 ⫾1.2%

after euglycemia, P⬍0.0001); total cho-

lesterol decreased (5.40 ⫾1.09 before vs.

4.50 ⫾1.10 mmol/l after euglycemia, P⬍

0.05); and triglycerides tended to de-

crease (4.34 ⫾3.32 before vs. 1.91 ⫾

0.69 mmol/l after euglycemia, P⫽0.06).

These results are described in detail else-

where (17). Plasma total homocysteine

levels were similar before and after im-

proved insulin sensitivity (total homocys-

teine levels 8.5 ⫾2.4 before vs. 9.2 ⫾3.5

␮mol/l after euglycemia, P⫽0.5). Urine

creatinine excretion was similar before

and after euglycemia (15.0 ⫾4.8 before

vs. 12 ⫾4.3 ␮mol/24 h after euglycemia).

This intervention study shows that

despite amelioration of hyperglycemia-

induced insulin resistance and improved

metabolic control, plasma total homocys-

teine concentrations did not decrease.

Homocysteine levels, measured in these

patients with type 2 diabetes, were actu-

ally within the normal range of healthy

subjects (2).

Our study is the ﬁrst intervention

study to investigate the relation between

insulin resistance and plasma total homo-

cysteine levels. To date only cross-

sectional studies have been performed

(13–15), which are more susceptible for

confounding by other metabolic factors.

Also in contrast to previous reports

(8,9,11,12,14,15), we studied insulin

sensitivity by clamp, the gold standard.

We conclude that amelioration of hy-

perglycemia-induced insulin resistance

does not change plasma total homocys-

teine levels. These data refute the hypoth-

esis that homocysteine levels are

inﬂuenced by insulin resistance and by

the degree of metabolic control, at least in

the patients we studied. Therefore, these

results do not support elevated homocys-

teine levels as an explanation for the link

between insulin resistance and the in-

creased risk of cardiovascular disease in

type 2 diabetes.

Acknowledgments—This study was sup-

ported by a grant from the Dutch Diabetes

Foundation. H.J.B. is an established investiga-

tor of the Netherlands Heart Foundation

(D97.021).

MARIE-JOSE J. POUWELS,

1,2

MARTIN DEN HEIJER,

MD, PHD

HENK J. BLOM,

PHD

CEES J. TACK,

MD, PHD

1,2

ADR. HERMUS,

MD, PHD

From the

Department of Endocrinology, University

Medical Center Nijmegen, the Netherlands; the

De-

partment of General Internal Medicine, University

Medical Center Nijmegen, the Netherlands; and the

Laboratory of Pediatrics and Neurology, University

Medical Center Nijmegen, the Netherlands.

Address correspondence to Marie-Jose Pouwels,

MD, Division of General Internal Medicine, Medisch

Spectrum Twente, P.O. Box 50000, 7500 KA En-

schede, The Netherlands. E-mail: m.j.pouwels@

club.tip.nl.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Haffner SM, Lehto S, Ro¨nnemaa T,

Pyo¨ra¨la¨K, Laakso M: Mortality from cor-

onary heart disease in subjects with type 2

diabetes and in nondiabetic subjects with

and without prior myocardial infarction.

N Engl J Med 339:229–234, 1998

2. Welch GN, Loscalzo J: Homocysteine and

atherothrombosis.NEngl JMed338:1042–

1050, 1998

3. Hoogeveen EK, Kostense PJ, Beks PJ,

Mackaay AJC, Jakobs C, Bouter LM,

Heine RJ, Stehouwer CD: Hyperhomocys-

teinemia is associated with an increased

risk of cardiovascular disease, especially

in non-insulin-dependent diabetes melli-

tus: a population-based study. Arterioscler

Thromb Vasc Biol 18:133–138, 1998

4. Hoogeveen EK, Kostense PJ, Jakobs C,

Dekker JM, Nijpels G, Heine RJ, Bouter

LM, Stehouwer CD: Hyperhomocysteine-

mia increases risk of death, especially in

type 2 diabetes: 5-year follow-up of the

hoorn study. Circulation 101:1506–1511,

2000

5. Rossetti L: Glucose toxicity. Clin Invest

Med 18:255–260, 1995

6. Fonseca V, Dicker-Brown A, Ranganathan

S, Song W, Barnard RJ, Fink L, Kern PA:

Effects of a high-fat-sucrose diet on en-

zymes in homocysteine metabolism in the

rat. Metabolism 49:736–741, 2000

7. Giltay EJ, Hoogeveen EK, Elber JM,

Gooren LJ, Asscheman H, Stehouwer CE:

Insulin resistance is associated with ele-

vated plasma total homocysteine levels in

healthy, non-obese subjects. Atherosclero-

sis 139:197–198, 1998

8. Meigs JB, Jacques PF, Selhub J, Singer DE,

Nathan DM, Rifai N, D’Agostino RB Sr,

Wilson PW: Fasting plasma homocys-

teine levels in the insulin resistance syn-

drome: the Framingham offspring study.

Diabetes Care 24:1403–1410, 2001

9. De Pergola G, Pannacciulli n, Zamboni M,

Minenna A, Brocco G, SciarafﬁaM,

Bosello, Giorgino R: Homocysteine

plasma levels are independently associ-

ated with insulin resistance in normal

weight, overweight and obese pre-meno-

pausal women. Diabetes Nutr Metab 14:

253–258, 2001

10. Abbasi F, Facchini F, Humphreys MH,

Reaven GM: Plasma homocysteine con-

centrations in healthy volunteers are not

related to differences in insulin-mediated

glucose disposal. Atherosclerosis 146:175–

178, 1999

11. Bar-on H, Kidron M, Friedlander Y, Ben-

Yehuda A, Selhub J, Rosenberg IH, Fried-

man G: Plasma total homocysteine levels

in subjects with hyperinsulinemia. J Intern

Med 247:287–294, 2000

12. Godsland IF, Rosankiewicz JR, Proudler

AJ, Johnston DG: Plasma total homocys-

teine concentrations are unrelated to in-

sulin sensitivity and components of the

metabolic syndrome in healthy men.

J Clin Endocrinol Metab 86:719–723, 2001

13. Emoto M, Kanda H, Shoji T, Kawagishi T,

Komatsu M, Mori K, Tahara H, Ishimura

E, Inaba M, Okuno Y, Nishizawa Y: Im-

pact of insulin resistance and nephropa-

thy on homocysteine in type 2 diabetes.

Diabetes Care 24:533–538, 2001

14. Buysschaert M, Dramais AS, Wallemacq

PE, Hermans MP: Hyperhomocysteine-

mia in type 2 diabetes: relationship to

macroangiopathy, nephropathy, and in-

sulin resistance. Diabetes Care 23:1816–

1822, 2000

15. Drzewoski J, Czupryniak L, Chwatko G,

Bald E: Total plasma homocysteine and

insulin levels in type 2 diabetic patients

with secondary failure to oral agents. Di-

abetes Care 22:2097–2098, 1999

16. Chan NN: Homocysteine and insulin lev-

els in type 2 diabetic patients (Letter). Di-

abetes Care 23:1041, 2000

17. Pouwels MJ, Tack CJ, Hermus AR, Lutter-

man JA: Treatment with intravenous

insulin followed by continuous subcuta-

neous insulin infusion improves glycemic

control in severely resistant type 2 dia-

betic patients. Diabet Med 20:76–79,

2003

18. Fiskerstrand T, Refsum H, Kvalheim G,

Ueland PM: Homocysteine and other thi-

ols in plasma and urine: automated deter-

mination and sample stability. Clin Chem

39:263–271, 1993

Letters

1638 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

19. te Poele-Pothoff MT, van der Berg M,

Franken DG, Boers GH, Jacobs C, de

Kroon IF, Eskes TK, Trijbels JM, Blom HJ:

Three different methods for the determi-

nation of total homocysteine in plasma.

Ann Clin Biochem 32:218–220, 1995

The True Overall

Mortality of

Unselected

Nephropathic Type 2

Diabetic Patients

With Mild to

Moderate Renal

Insufﬁciency

espite signiﬁcant improvements in

therapy over the last decade, the

prognosis of nephropathic type 2

diabetic patients continues to be poor (1).

In recent antihypertensive intervention

studies performed in type 2 diabetic pa-

tients with overt nephropathy, the mean

5-year mortality rate was ⬃20%. In the

irbesartan study published by Lewis et al.

(2), the mean death rate after 54 months

ranged from 15 (irbesartan group) to

16.3% (placebo group). Similar data were

obtained in the losartan study published

by Brenner et al. (3). These results might

lead to false conclusions concerning the

overall mortality of all unselected type 2

diabetic patients with overt nephropathy.

In the irbesartan study, major excretion

criteria were congestive heart failure

(New York Heart Association [NYHA]

class III or worse) and cardiovascular

events within the last 3 months.

In accordance with the irbesartan

study, in 1996 we recruited all nephro-

pathic type 2 diabetic patients who were

treated at that time in our hospital and/or

in the outpatient care unit and who ful-

ﬁlled the inclusion criteria of the study

(serum creatinine 1.0–3.0 mg/dl in fe-

male and 1.3- 3.0 mg/dl in male patients,

as well as protein excretion ⱖ1.0 g/24-h

urine), without considering the usual ex-

clusion criteria. We recruited 46 nephro-

pathic type 2 diabetic patients (aged 61 ⫾

8 years, 26 women and 20 men, diabetes

duration 17 ⫾4 years), and we prospec-

tively studied their progression of diabetic

nephropathy and the 5-year mortality rate.

Cardiovascular disease was present in

43%, while 28% had heart failure (NYHA

class III and higher). Of patients, 36% had

exclusion criteria according to the irbe-

sartan study. All patients underwent eval-

uations in our outpatient unit (n⫽24) or

at the ofﬁces of their general physicians in

at least 6-month intervals over an obser-

vation period of 5 years. The following

parameters were routinely measured: cre-

atinine in serum and 24-h urine, creati-

nine clearance (calculated), protein in

24-h urine, HbA

, and blood pressure.

The three end points of the study were

doubling of serum creatinine levels, end-

stage renal disease, or death. Patients were

also divided into those with and without

heart failure, and 5-year mortality rates

were assessed in both groups.

Serum creatinine had doubled in 37%

of patients over 60 months, compared

with 22% within 54 months in the irbe-

sartan study. The percentage of patients

who reached end-stage renal disease was

26% in our patients and 16.7% in the

irbesartan study. During the period of ob-

servation, mean HbA

level in our pa-

tients was 7.6 ⫾1.0%, while their mean

blood pressure was 139 ⫾11/80 ⫾8

mmHg. In the irbesartan study, the mean

blood pressure at visits after baseline was

similar, with 140/77 mmHg in the irbe-

sartan group and 144/80 mmHg in the

placebo group.

The overall 5-year mortality of all of

our diabetic patients was 33%. Five-year

mortality rates were 57% in patients with

heart failure (NYHA III or IV) and 22%

(P⬍0.05) in those without heart failure.

In comparison with the data of the inter-

vention studies, the mortality of our ne-

phropathic type 2 diabetic patients

without heart failure was similar to that of

the selected patients in the irbesartan

study. In contrast, survival was signiﬁ-

cantly lower in patients with heart failure,

and as a consequence, the overall mortal-

ity of all unselected type 2 diabetic pa-

tients with overt nephropathy was also

higher than that registered in the inter-

vention studies. There were no signiﬁcant

differences at baseline visit in creatinine

clearance (139 ⫾31 vs. 134 ⫾32 ml 䡠

min

⫺1

䡠1.73 m

⫺2

) and urinary protein

(2.4 ⫾1.3 vs. 2.3 ⫾1.2 g/24 h) in the

groups with and without chronic heart

failure, respectively. The mean blood

pressure during the observation period

was approximately the same in both

groups (140 ⫾12 vs. 138 ⫾11 mmHg).

Twenty-eight percent of the patients with

and 25% of those without heart failure

reached end-stage renal disease. Our re-

sults are in agreement with those of Short

(4), who found, at a mean follow-up of

5.3 years, a mortality rate of 37% in ne-

phropathic type 2 diabetic patients com-

pared with just 8% in a nonnephropathic

group. Multivariate Cox regression anal-

yses conﬁrmed a ﬁvefold excess risk for

cardiovascular mortality in type 2 diabetic

patients with overt nephropathy. Angio-

tensin II blockers have shown effective

cardiovascular protection in type 2 dia-

betic patients (5).

In conclusion, the real mortality rate

of all unselected type 2 diabetic patients

with overt nephropathy and mild-to-

moderate renal insufﬁciency is much

higher than the mortality of nephropathic

type 2 diabetic patients selected for anti-

hypertensive studies. The difference be-

tween outcome in studies versus “real life”

is mainly due to the fact that patients with

heart failure are excluded from studies,

but obviously present in a nonstudy

situation.

GEORG BIESENBACH,

OTMAR JANKO,

From the 2nd Department of Medicine, General

Hospital, Linz, Austria.

Address correspondence to Georg Biesenbach,

MD, 2nd Department of Medicine, General Hospi-

tal, Krankenhausstrasse 9, 4020 Linz, Austria. E-

mail: georg.biesenbach@akh.linz.at.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Ismail N, Becker B, Strzelczyk P, Ritz E:

Renal disease and hypertension in non-

insulin-dependent diabetes mellitus. Kid-

ney Int 55:1–28, 1999

2. Lewis EJ, Hunsicker LG, Bain RP, Clarke

WR, Berl T, Pohl MA, Lewis JB, Ritz E,

Atkins RC, Rhode R: Renoprotective effect

of the angiotensin-receptor antagonist

irbesartan in patients with nephropathy

due to type 2 diabetes. N Engl J Med 345:

851–860, 2001

3. Brenner BM, Cooper ME, Dick de Zeeuw,

Keane WF, Mitch WE, Parving HH, Re-

muzzi G, Snapinn SM, Zhang Z, Shahnaz

S, for the RENAAL Study Investigators:

Effects of losartan on renal and cardiovas-

cular outcomes in patients with type 2 di-

abetes and nephropathy. N Engl J Med

345:861–869, 2001

4. Short R: Natural history and prognostic

factors: nephropathy in type 2 diabetes. Q

J Med 95:371–377, 2002

5. Deferrari G, Ravers M, Deferrari L, Vettor-

etti S, Ratto E, Parodi D: Renal and cardio-

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1639

vascular protection in type 2 diabetes

mellitus: angiotensin II receptor blockers.

J Am Soc Nephrol 13 (Suppl. 3):224–229,

2002

HLA-DR-DQ

Haplotype in Rapid-

Onset Type 1

Diabetes in

Japanese

ype 1 diabetes is characterized by

the presence of insulitis and auto-

antibodies, at least in the earlier

course of the disease. It is well known that

HLA locus confers strong genetic suscep-

tibility to the disease. Imagawa et al. (1)

described a unique subtype of type 1 di-

abetes that shows a rapid and fulminant

onset of the disease but lacks evidence of

insulitis and autoantibodies. Although

the etiology of this unique subtype of type

1 diabetes is unclear at present, Tanaka et

al. (2) very recently reported that speciﬁc

HLA-DQ genotypes may be associated

with this subtype of type 1 diabetes. Here,

we report clinical characteristics of sub-

jects with type 1 diabetes of a rapid and

fulminant onset who were collected from

the registration of type 1 diabetes in our

institution.

We identiﬁed 19 type 1 diabetic sub-

jects showing an acute and abrupt onset

of diabetes (duration of hyperglycemic

symptoms 6 ⫾3 days [mean ⫾SD],

range 1–10) accompanied by low HbA

levels (6.8 ⫾0.9%, 5.4–8.2). The sub-

jects consisted of 11 men and 8 women

aged 38 ⫾17 years (range 16–66) with

BMI of 22.2 ⫾3.2 kg/m

(17.1–27.7).

Family history of diabetes (possibly type 2

diabetes) was recorded in three subjects

(16%). Among eight women, four mani-

fested diabetes during gestation (n⫽2) or

just after delivery (n⫽2). History of pre-

ceding upper respiratory infection was

noted in nine subjects (47%). Despite

short duration of subjective symptoms

and low HbA

levels, their initial concen-

trations of plasma glucose was as high as

34.1 ⫾13.2 mmol/l (range 13.6–72.4),

indicating a fulminant onset of diabetes.

In fact, 14 subjects showed diabetic keto-

acidosis at their initial presentation (arte-

rial pH 7.14 ⫾0.17, 6.80–7.34), and the

remaining ﬁve had marked ketosis and

hyperglycemia. GAD autoantibodies,

measured by commercial RIA kits (Cos-

mic Corporation, Tokyo) (3), were nega-

tive in 18 of the 19 subjects (⬍1.5 units/

ml), but a single subject showed a weakly

positive level (5.6 units/ml). Insulinoma-

associated protein 2 (IA2) autoantibody

and anti-thyroperoxidase and -thyroglob-

ulin antibodies were negative in all sub-

jects. Fasting serum level and urinary

excretion of C-peptide were decreased to

0.10 ⫾0.07 nmol/l (0–0.26) and 2.2 ⫾

2.1 ␮mol/day (0.2–6.9), respectively,

and they did not show any signs of remis-

sion of insulin dependency during the ob-

servation period.

As for HLA status, HLA-DRB1-DQB1

genotypes were determined and com-

pared with those in autoantibody (GAD

and/or IA2 autoantibodies)-positive sub-

jects with signiﬁcant type 1A diabetes

(n⫽87) derived from our registration.

Frequency of HLA-DRB1*0405-

DQB1*0401 haplotype in a homozygous

manner was signiﬁcantly higher in the

rapid-onset group (5 of 19, 26%) than the

type 1A group (4 of 87, 5%, P⫽0.009 by

Fisher’s exact probability). The frequency

was also far higher than that of nondia-

betic Japanese subjects reported in the lit-

erature (1 of 84, P⬍0.001 [2] and 6 of

157, P⫽0.003 [4]). Frequency of other

susceptible haplotypes (DRB1*0405-

DQB1*0401 in a heterozygous manner

and DRB1*0901-DQB1*0303 in a ho-

mozygous manner) (4) did not differ be-

tween the rapid-onset group and type 1A

group (data not shown).

In our experience, homozygous HLA-

DRB1*0405-DQB1*0401 haplotype con-

fers genetic susceptibility to rapid-onset

type 1 diabetes in Japanese, consistent

with the recent report by Tanaka et al. (2).

Frequency of GAD autoantibodies was ac-

tually low in this subgroup of type 1 dia-

betes (1,2), but a single subject showed a

positive result at the onset of diabetes.

There was also a case report in which

GAD autoantibodies turned positive 1

year after the fulminant onset of type 1

diabetes (5). Tanaka et al. (6) also dem-

onstrated presence of insulitis in a case of

rapid-onset type 1 diabetes. Collectively,

it is suggested that rapid-onset type 1 di-

abetes could share autoimmune etiology

with type 1A diabetes (7). Of interest is

that half of the female patients developed

diabetes during or just after pregnancy.

There have been many studies reporting

altered immune response during preg-

nancy (8,9), and pregnancy is considered

to be a precipitating factor for type 1 dia-

betes (10). We hypothesize that altered

immune response to unknown stimulus

may be causative of rapid and aggressive

␤-cell loss, together with the genetic sus-

ceptibility, at least in part, determined by

HLA-DR-DQ haplotypes.

TOMOATSU NAKAMURA,

SHOICHIRO NAGASAKA,

IKUYO KUSAKA,

TOSHIMITSU YATAGAI,

JIANMEI YANG,

SHUN ISHIBASHI,

From the Division of Endocrinology and Metabo-

lism, Jichi Medical School, Tochigi, Japan.

Address correspondence to Shoichiro Nagasaka,

MD, Division of Endocrinology and Metabolism, Ji-

chi Medical School, Yakushiji 3311-1, Minami-

kawachi, Tochigi 329-0498, Japan. E-mail:

sngsk@jichi.ac.jp.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Imagawa A, Hanafusa T, Miyagawa J, Mat-

suzawa Y: A novel subtype of type 1 dia-

betes mellitus characterized by a rapid

onset and an absence of diabetes-related

antibodies. N Engl J Med 342:301–307,

2000

2. Tanaka S, Kobayashi T, Nakanishi K,

Koyama R, Okubo M, Murase T, Odawara

M, Inoko H: Association of HLA-DQ ge-

notype in autoantibody-negative and rap-

id-onset type 1 diabetes. Diabetes Care 25:

2302–2307, 2002

3. Akamine H, Komiya I, Shimabukuro T,

Asawa T, Tanaka H, Yagi N, Taira T, Na-

gata K, Arakaki K, Wakugami T, Takasu

N, Powell MJ, Furmaniak J, Smith BR:

High prevalence of GAD65 (and IA-2) an-

tibodies in Japanese IDDM patients by a

new immunoprecipitation assay based on

recombinant human GAD65. Diabet Med

14:778–784, 1997

4. Kawabata Y, Ikegami H, Kawaguchi Y, Fu-

jisawa T, Shintani M, Ono M, Nishino M,

Uchigata Y, Lee I, Ogihara T: Asian-spe-

ciﬁc HLA haplotypes reveal heterogeneity

of the contribution of HLA-DR and -DQ

haplotypes to susceptibility to type 1 dia-

betes. Diabetes 51:545–551, 2002

5. Shimada A, Oikawa Y, Shigihara T, Senda

T, Kodama K: A case of fulminant type 1

diabetes with strong evidence of autoim-

munity (Letter). Diabetes Care 25:1482–

1483, 2002

6. Tanaka S, Kobayashi T, Momotsu T: A

novel subtype of type 1 diabetes mellitus.

N Engl J Med 342:1835–1837, 2000

7. Lernmark A: Rapid-onset type 1 diabetes

with pancreatic excorine dysfunction.

N Engl J Med 342:344–345, 2000

Letters

1640 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

8. Whitacre CC, Reingold SC, O’Looney PA:

A gender gap in autoimmunity. Science

283: 1277–1278

9. Elenkov IJ, Wilder RL, Bakalov VK, Link

AA, Dimitrov MA, Fisher S, Crane M,

Kanik KS, Chrousos GP: IL-12, TNF-␣,

and hormonal changes during late preg-

nancy and early postpartum: implications

for autoimmune disease activity during

these times. J Clin Endocrinol Metab 86:

4933–4938, 2001

10. Buschard K, Buch I, Molsted-Pedersen L,

Hougaad P, Kuhl C: Increased incidence

of true type 1 diabetes acquired during

pregnancy. BMJ 294:275–279, 1987

Sex, Diabetes, and

Stroke After Carotid

Endarterectomy

hile diabetes is associated with

increased risk of perioperative

stroke among subjects who un-

dergo carotid endarterectomy (1–3),

women and minorities have been gener-

ally underrepresented in such trials (4).

The North American Symptomatic Ca-

rotid Endarterectomy Trial (NASCET)

was a randomized study of 2,885 eligible

patients with carotid stenosis ⬍70% who

received either medical (n⫽1,449) or

surgical therapy (n⫽1,436 patients) (3).

The median study age was 66 years, and

93% of study subjects were white. Ap-

proximately 30% of NASCET subjects

were female, and ⬃22% had a history of

diabetes, which was almost always type 2

diabetes (3). The study showed that end-

arterectomy safely and efﬁcaciously re-

duced the risk of stroke in symptomatic

patients with higher-grade carotid arterial

stenosis when compared with medical

therapy (3). Multivariate analysis indi-

cated an increased risk of perioperative

stroke in subjects with a history of diabe-

tes (relative risk [RR] 2.0, 95% CI 1.2–

3.1) (3). Because NASCET had a relatively

high proportion of both female and dia-

betic subjects, we addressed the hypoth-

esis that there would be differences

between diabetic men and women with

respect to the risk of peri- and postoper-

ative complications. Subjects were subdi-

vided by surgical or medical therapy and

then subdivided again by sex and diabetes

history. End point data regarding strokes

and deaths were collected.

Diabetic men had a signiﬁcantly in-

creased risk of stroke or death 30 days

postendarterectomy compared with non-

diabetic men (RR 2.3, 95% CI 1.4–3.7),

with no difference between diabetic and

nondiabetic women (1.1, 0.5–2.4). Fur-

thermore, diabetic men had a signiﬁ-

cantly increased risk of stroke 3 years

postendarterectomy compared with non-

diabetic men (RR 1.9, 95% CI 1.3–2.8),

again with no difference between diabetic

and nondiabetic women (1.0, 95% CI

0.6–1.9). However, at 3 years postran-

domization, diabetic men who received

medical treatment had no increased risk

of stroke compared with nondiabetic men

(RR 0.7, 0.5–1.1). In contrast, at 3 years

postrandomization, diabetic women who

received medical treatment had a signiﬁ-

cantly increased risk of stroke compared

with nondiabetic women (1.7, 1.1–2.8).

The results support the emerging im-

pression of sex-related differences among

diabetic subjects with respect to vascular

disease (5,6). Perioperative neurological

morbidity of carotid endarterectomy is

mainly due to embolization at the time of

clamping or clamp release, ischemia dur-

ing clamping of extracranial arteries, and

intracerebral hemorrhage (7). There has

been a steady decrease in postoperative

morbidity among patients who undergo

carotid endarterectomy as a result of both

improved surgical technique and patient

selection (7). Baseline clinical attributes

are also helpful predictors of postopera-

tive mortality and morbidity from stroke

(3). Our analysis suggests that among di-

abetic subjects who undergo carotid end-

arterectomy, there may be sex-related

differences in postoperative neurological

morbidity. As with any post hoc subgroup

analysis, the results should be interpreted

cautiously and must be replicated in other

studies. However, if these ﬁndings are

conﬁrmed, it might be appropriate to

consider sex and diabetes status in strate-

gies to optimize the management of pa-

tients with carotid disease (8).

ROBERT A. HEGELE,

MD, FRCPC

MICHAEL ELIASZIW,

PHD

MASSIMILIANO DEANGELIS,

From the

Robarts Research Institute, London, On-

tario, Canada; the

Department of Community

Health Sciences, Faculty of Medicine, University of

Calgary, Calgary, Alberta, Canada; and the

Depart-

ment of Internal Medicine, Endocrinology and Me-

tabolism Section, University of Perugia, Perugia, Italy.

Address correspondence to Dr. Robert Hegele,

Robarts Research Institute, London, Ontario N6A

5K8, Canada. E-mail: hegele@robarts.ca.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Salenius JP, Harju E, Riekkinen H: Early

cerebral complications in carotid endar-

terectomy: risk factors. J Cardiovasc Surg

31:162–167, 1990

2. Akbari CM, Pomposelli FBJ, Gibbons

GW, Campbell DR, Freeman DV, LoGerfo

FW: Diabetes mellitus: a risk factor for

carotid endarterectomy? J Vasc Surg 25:

1070–1075, 1997

3. Barnett HJ, Taylor DW, Eliasziw M, Fox

AJ, Ferguson GG, Haynes RB, Rankin RN,

Clagett GP, Hachinski VC, Sackett DL:

Beneﬁt of carotid endarterectomy in pa-

tients with symptomatic moderate or severe

stenosis: North American Symptomatic

Carotid Endarterectomy Trial Collabora-

tors. N Engl J Med 339:1415–1425, 1998

4. Rigdon EE: Racial and sex differences in

outcome after carotid endarterectomy.

Am Surg 64:527–530, 1998

5. Anderson KM, Odell PM, Wilson PW,

Kannel WB: Cardiovascular disease risk

proﬁles. Am Heart J 121:293–298, 1991

6. Wilson PW: Diabetes mellitus and coro-

nary heart disease. Am J Kidney Dis 32:

S89–S100, 1998

7. Riles TS, Imparato AM, Jacobowitz GR,

Lamparello PJ, Giangola G, Adelman MA,

Landis R: The cause of perioperative

stroke after carotid endarterectomy. J Vasc

Surg 19:206–214, 1994

8. Barnett HJ, Meldrum HE, Eliasziw M,

North American Symptomatic Carotid

Endarterectomy Trial (NASCET) collabo-

rators: The appropriate use of carotid

endarterectomy.CMAJ 166:1169–1179,

2002

The Diabetes

Empowerment Scale–

Short Form (DES-SF)

n 2000 we developed the Diabetes Em-

powerment Scale (DES) to measure the

psychosocial self-efﬁcacy of people

with diabetes. The original questionnaire

contained 37 items representing eight

conceptual dimensions (i.e., assessing the

need for change, developing a plan, over-

coming barriers, asking for support, sup-

porting oneself, coping with emotion,

motivating oneself, and making diabetes

care choices appropriate for one’s priori-

ties and circumstances). Using factor

analyses the questionnaire was reduced to

the current 28-item DES (␣⫽0.96) con-

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1641

taining three subscales (1). The three sub-

scales are: 1) managing the psychosocial

aspects of diabetes with 9 items (␣⫽

0.93), 2) assessing dissatisfaction and

readiness to change with 9 items (␣⫽

0.81), and 3) setting and achieving goals

with 10 items (␣⫽0.91). In addition to

providing an overall assessment of dia-

betes-related psychosocial self-efﬁcacy,

the three subscales of the DES allow for an

examination of its underlying components.

To allow for a brief overall assessment

of diabetes-related psychosocial self-

efﬁcacy, we developed an eight-item short

form of the DES (the DES-SF). The

DES-SF was created by choosing the item

from the remaining 28 items with highest

item to subscale correlation from each of

the original eight conceptual domains.

The reliability of the DES-SF using the

original dataset was ␣⫽0.85. We have

subsequently administered the DES-SF to

229 subjects in a new study. The reliabil-

ity of the DES-SF using the data from the

new sample was ␣⫽0.84. The content

validity of the DES-SF was supported in

the new study by the fact that both

DES-SF scores and HbA

levels changed

in a positive direction after the 229 sub-

jects completed a 6-week problem-based

patient education program (2). The

change in DES-SF scores and HbA

levels

were not correlated, suggesting that these

two measures vary independently.

These data provide preliminary evi-

dence that the DES-SF is a valid and reli-

able measure of overall diabetes-related

psychosocial self-efﬁcacy. The DES and

the DES-SF, scoring information, and

permission to use them can be down-

loaded from the Michigan Diabetes Re-

search and Training Center web site at:

www.med.umich.edu/mdrtc.

ROBERT M. ANDERSON,

EDD

JAMES T. FITZGERALD,

PHD

LARRY D. GRUPPEN,

PHD

MARTHA M. FUNNELL,

MS, RN, CDE

MARY S. OH,

From the

University of Michigan, DRTC, Depart-

ment of Medical Education, Ann Arbor, Michigan;

and the

University of Michigan, Diabetes Research

and Training Center, Ann Arbor, Michigan.

Address correspondence to Robert M. Anderson,

Department of Medical Education, G1100 Towsley

Center, 1500 East Medical Center Dr., Ann Arbor,

MI 48109-0201. E-mail: boba@umich.edu.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Anderson RM, Funnell MM, Fitzgerald JT,

Marrero DG: The diabetes empowerment

scale: a measure of psychosocial self-efﬁ-

cacy. Diabetes Care 23:739–743, 2000

2. Anderson RM, Funnell MM, Nwankwo R,

Gillard ML, Fitzgerald JT, Oh M: Eval-

uation of a problem-based, culturally

speciﬁc, patient education program for

African Americans with diabetes (Ab-

stract). Diabetes 50 (Suppl. 2):A195, 2001

Increased Oxidative

Stress Is Associated

With Elevated

Plasma Levels of

Adrenomedullin in

Hypertensive

Patients With Type 2

Diabetes

revious studies have demonstrated

that oxidative stress is associated not

only with hyperglycemia and hyper-

tension but also with the metabolic syn-

drome and cadiovascular disease (1–3).

Adrenomedullin (AM) is a novel va-

sorelaxant peptide isolated from human

pheochromocytoma (4,5). Oxidative

stress enhances AM production from en-

dothelium and vascular smooth muscle

cells in vitro (6,7). However, whether ox-

idative stress is associated with circulating

levels of AM in vivo is unknown. In the

present study, we evaluated the relation-

ship between the plasma levels of 8-

epi-prostaglandinF2␣(8-epi-PGF2␣;

currently regarded as the most reliable

marker for the assessment of oxidative

stress in humans) (8,9) and AM in normal

subjects and hypertensive patients with

type 2 diabetes.

This study comprised 17 hyperten-

sive patients with type 2 diabetes (15 men

and 2 women, age 47.3 ⫾3.0 years

[mean ⫾SE], BMI 23.2 ⫾0.8 kg/m

, fast-

ing plasma glucose 8.2 ⫾0.5 mmol/l,

HbA

9.9 ⫾0.5%, fasting serum insulin

30.6 ⫾3.0 pmol/l, systolic blood pres-

sure 150.2 ⫾3.4 mmHg, and diastolic

blood pressure 76.5 ⫾2.2 mmHg) and

18 normal subjects (17 men and 1

women, age 43.0 ⫾1.9 years, BMI 23.8 ⫾

0.4 kg/m

, fasting plasma glucose 5.1 ⫾

0.2 mmol/l, fasting serum insulin 31.2 ⫾

3.6 pmol/l, systolic blood pressure

126.8 ⫾2.2 mmHg, and diastolic blood

pressure 79.8 ⫾1.6 mmHg). All subjects

were nonsmokers. Hypertensive patients

with type 2 diabetes were diagnosed at a

local clinic 3.0 ⫾0.2 years before the be-

ginning of this study. All patients were

treated with diet (1,440–1,720 kcal/day

and sodium restriction 304 mmol/day)

and exercise (walking 10,000 steps/day);

none of them were receiving any kind of

drug. Their blood glucose and blood pres-

sure were in good control (HbA

ⱕ6.5%,

systolic blood pressure ⬍140 mmHg, and

diastolic blood pressure ⬍90 mmHg)

during the initial several months. How-

ever, thereafter the blood glucose and

blood pressure of the patients gradually

increased (HbA

ⱖ8%, systolic blood

pressure ⱖ140 mmHg, and diastolic

blood pressure ⱖ90 mmHg) because of

overeating and inactivity. They were re-

ferred to our clinical department for con-

trol of hyperglycemia and hypertension.

None of them had evidence of micro- or

macroangiopathy. Informed consent was

obtained from all subjects before the be-

ginning of the study. Plasma levels of free

8-epi-PGF2␣were measured using a

commercially available enzyme immuno-

assay kit (Cayman Chemical, Ann Arbor,

MI). The detection limit of this assay was

1.5 pg/ml, and the intra- and interassay

coefﬁcients of variation were 7.5 and

9.2%, respectively. AM in plasma samples

was measured using a commercially avail-

able immunoradiometric assay kit

(Shionogi Pharmaceuticals, Osaka, Ja-

pan). The detection limit of this assay was

2 fmol/ml, and the intra- and interassay

coefﬁcients of variation were 7.0 and

6.9%. Serum insulin was measured using

an immunoradiometric assay kit (Insulin

Riabead II kit; Dainabot, Tokyo). The in-

tra- and interassay coefﬁcients of varia-

tion of the assay were 1.9 and 2.0%. In

addition, we measured blood pressure in

supine position after a 5-min rest.

Both plasma levels of 8-epi-PGF2␣

and AM were signiﬁcantly increased in

hypertensive patients with type 2 diabetes

compared with normal subjects (8-epi-

PGF2␣48.6 ⫾8.6 vs. 11.9 ⫾1.3 pg/ml,

P⬍0.05; AM 14.8 ⫾0.7 vs. 12.4 ⫾0.2

fmol/ml, P⬍0.02). The plasma levels of

8-epi-PGF2␣were proportionally corre-

lated with AM (r⫽0.696, P⬍0.01) in

only hypertensive patients with type 2 di-

abetes. Signiﬁcant positive correlations

were observed between plasma levels of

Letters

1642 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

8-epi-PGF2␣(r⫽0.540, P⬍0.05) or

AM (r⫽0.875, P⬍0.001) and systolic

blood pressure in patients with type 2 di-

abetes.

This is the ﬁrst report that demon-

strated a relationship between oxidative

stress and AM in vivo. The mechanism by

which plasma levels of 8-epi-PGF2␣cor-

relate with AM and the cellular source of

AM remain unknown. Previous studies

have shown that oxidative stress stimu-

lates secretion of AM from endothelium

and vascular smooth muscle cells and that

AM mRNA expression is increased by ac-

tivation of the nuclear factor-␬B pathway

(6,7). Increased AM secretion from endo-

thelium and vascular smooth muscle may

inﬂuence the plasma levels of AM. On the

other hand, Shimosawa et al. (10) re-

ported that endogenous AM may protect

from organ damage by inhibiting oxida-

tive stress production. Increased AM lev-

els may compensate for oxidative stress–

induced vasoconstriction and thus may

play a protective role against organ injury.

Adrenal medulla and pancreatic islets

may also be a source of AM (4,11). Fur-

ther study is needed to clarify the source

of plasma AM and the mechanism of cor-

relation between oxidative stress and AM.

In conclusion, there was a signiﬁcant

positive correlation between increased

oxidative stress and elevated plasma lev-

els of AM in hypertensive patients with

type 2 diabetes. Enhanced oxidative stress

may regulate the plasma levels of AM in

hypertensive patients with type 2 diabetes.

AKIRA KATSUKI,

YASUHIRO SUMIDA,

HIDEKI URAKAWA,

ESTEBAN C. GABAZZA,

NORIKO MARUYAMA,

KOHEI MORIOKA,

NAGAKO KITAGAWA,

YASUKO HORI,

KANAME NAKATANI,

YUTAKA YANO,

YUKIHIKO ADACHI,

From the

Third Department of Internal Medicine,

Mie University School of Medicine, Mie, Japan; and

the

Department of Laboratory Medicine, Mie Uni-

versity School of Medicine, Mie, Japan.

Address correspondence to A. Katsuki, MD,

Third Department of Internal Medicine, Mie Univer-

sity School of Medicine, 2-174 Edobashi, Tsu, Mie

514-8507, Japan. E-mail: katuki-a@clin.medic.mie-

u.ac.jp.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Ceriello A, Bortolotti N, Motz E, Crescen-

tini A, Lizzio S, Russo A, Tonutti L,

Taboga C: Meal-generated oxidative stress

in type 2 diabetic patients. Diabetes Care

21:1529–1533, 1998

2. Yasunari K, Maeda K, Nakamura M, Yo-

shikawa J: Oxidative stress is a possible

link between blood pressure, blood glu-

cose, and C-reacting protein. Hypertension

39:777–780, 2002

3. Evans JL, Goldﬁne ID, Maddux BA, Grod-

sky GM: Oxidative stress and stress-acti-

vated signaling pathways: a unifying

hypothesis of type 2 diabetes. Endocr Rev

23:599–622, 2002

4. Kitamura K, Kangawa K, Kawamoto M,

Ichiki Y, Nakamura S, Matsuo H, Eto T:

Adrenomedullin: a novel hypotensive

peptide isolated from human pheochro-

mocytoma. Biochem Biophys Res Commun

192:553–560, 1993

5. Kinoshita H, Kato K, Kuroki M, Naka-

mura S, Kitamura K, Hisanaga S, Fuji-

moto S, Eto T: Plasma adrenomedullin

levels in patients with diabetes (Letter).

Diabetes Care 23:253–254, 2000

6. Chun T-H, Itoh H, Saito T, Yamahara K,

Doi K, Mori Y, Ogawa Y, Yamashita J,

Tanaka T, Inoue M, Masatsugu K, Sawada

N, Fukunaga Y, Nakao K: Oxidative stress

augments secretion of endothelium-de-

rived relaxing peptides, C-type natriuretic

peptide and adrenomedullin. J Hypertens

18:575–580, 2000

7. Ando K, Ito Y, Kumada M, Fujita T: Oxi-

dative stress increases adrenomedullin

mRNA levels in cultured rat vascular

smooth muscle cells. Hypertens Res 21:

187–191, 1998

8. Sampson MJ, Gopaul N, Davies IR,

Hughes DA, Carrier MJ: Plasma F2 iso-

prostanes, direct evidence of increased

free radical damage during acute hyper-

glycemia in type 2 diabetes. Diabetes Care

25:537–541, 2002

9. Gopaul NK, Manraj MD, Hebe A, Kwai

Yan SL, Johnston A, Carrier MJ, Anggard

EE: Oxidative stress could precede endo-

thelial dysfunction and insulin resistance

in Indian Mauritians with impaired glucose

metabolism. Diabetologia 44:706–712, 2001

10. Shimosawa T, Shibagaki Y, Ishibashi K,

Kitamura K, Kangawa K, Kato S, Ando K,

Fujita T: Adrenomedullin, an endogenous

peptide, counteracts cardiovascular dam-

age. Circulation 105:106–111, 2002

11. Katsuki A, Sumida Y, Gabazza EC, Mu-

rashima S, Urakawa H, Morioka K, Kita-

gawa N, Tanaka T, Araki-Sasaki R, Hori Y,

Nakatani K, Yano Y, Adachi Y: Acute hyper-

insulinemia is associated with increased

circulating levels of adrenomedullin in

patients with type 2 diabetes mellitus. Eur

J Endocrinol 147:71–75, 2002

Renin-Angiotensin

System Gene

Polymorphisms and

Retinopathy in

Chinese Patients

With Type 2

Diabetes

iabetic eye disease is the leading

cause of new cases of blindness in

many developed countries, for

which macular edema and proliferative

retinopathy are major causes of loss of vi-

sion (1). Genetic studies have previously

suggested that diabetic retinopathy is

partly determined by genetic factors (2).

Angiotensin II, produced locally within

the retina, has potent hemodynamic and

growth-promoting effects and stimulates

ocular neovascularization (3,4). The pres-

ence of retinopathy is associated with an

activated renin-angiotensin system (RAS)

(5), and inhibition of the RAS signiﬁ-

cantly reduced the progression of reti-

nopathy in nonhypertensive patients with

type 1 diabetes (6). The ACE gene inser-

tion/deletion (I/D) polymorphism has

been reported to be associated with reti-

nopathy in type 1 diabetic subjects, but

the ﬁndings are heterogeneous (7,8).

Only limited data are available describing

the angiotensinogen (AGT) M235T and

the angiotensin II type 1 receptor (AT

A1166C polymorphisms and retinopathy

(9,10).

In this study, we examined the asso-

ciations between these three polymor-

phisms and diabetic retinopathy in 827

Chinese type 2 diabetic patients (based on

1985 World Health Organization crite-

ria). All subjects gave written informed

consent and were of Han Chinese origin.

They were examined by the physicians or

ophthalmologists through dilated pupils.

Retinopathy was considered to be present

if there was one or more areas of hemor-

rhages, microaneurysms, cotton wool

spots, and/or laser coagulation scars re-

lated to diabetic retinopathy or history of

vitrectomy (11). The RAS genotype and

allele frequencies were identiﬁed using

PCR/restriction fragment–length poly-

morphism protocols (12) in 326 patients

with retinopathy and 501 diabetic pa-

tients without retinopathy who were

matched for age (59.8 ⫾11.4 vs. 60.4 ⫾

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1643

9.3 years), sex (44.2 female vs. 39.3%

male), disease duration (6.3 [range 5.6–

7.0] vs. 6.0 years [5.6–6.3]), and age of

onset of diabetes (53.2 ⫾9.7 vs. 51.9 ⫾

12.4 years). Of those with retinopathy,

66.6% had nonproliferative, 8.3% pre-

proliferative, and 8.9% proliferative dis-

ease and 16.2% had advanced eye disease.

Of the patients, 30.1% with retinopathy

had received laser therapy in at least one

eye, with 18.7, 29.2, 53.8, and 69.4% be-

ing treated in the groups with increasing

severity of retinopathy.

The allele frequencies were 32.7 and

32.1% for the ACE D allele, 16.1 and

13.8% for the AGT M allele, and 4.6 and

3.4% for the AT

R C allele in those with-

out and with retinopathy, respectively.

The genotype distributions for these gene

polymorphisms are described in Table 1.

No differences were identiﬁed in the ge-

notype or allele frequencies between the

groups either for the dichotomous classi-

ﬁcation of retinopathy or when the sever-

ity of the retinopathy was graded.

The ACE D allele has been associated

with increasing ACE levels (12). Subjects

with the angiotensinogen M235T poly-

morphism TT genotype have increased

serum AGT levels, while the AT

R 1166C

allele is associated with enhanced respon-

siveness to angiotensin II (13). The ACE

DD genotype was found to be associated

with the presence of proliferative retinop-

athy in a relatively small case-control

study of type 1 diabetic subjects (7). In

the current study, the analysis compar-

ing age-, sex-, and diabetes duration–

matched subjects found no evidence of

associations between the distribution of

genotypes or alleles of these RAS gene

polymorphisms and retinopathy. Our

data support the majority of the literature

regarding the ACE I/D gene polymor-

phism, which has reported no signiﬁcant

association with retinopathy in diabetic

patients (7,8).

G. NEIL THOMAS,

PHD

JULIAN A.J.H. CRITCHLEY,

FRCP

BRIAN TOMLINSON,

FRCP

VINCENT T.F. YEUNG,

DENNIS LAM,

CLIVE S. COCKRAM,

FRCP

JULIANA C.N. CHAN,

FRCP

From the

Department of Medicine and Therapeu-

tics, The Chinese University of Hong Kong, The

Prince of Wales Hospital, Shatin, Hong Kong; and

the

Department of Ophthalmology and Visual Sci-

ence, The Chinese University of Hong Kong, The

Prince of Wales Hospital, Shatin, Hong Kong.

Address correspondence to G. Neil Thomas, De-

partment of Medicine and Therapeutics, The Prince

of Wales Hospital, Shatin, Hong Kong. E-mail:

gneilthomas@yahoo.co.uk.

J.A.J.H.C. is deceased.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Klein R, Klein BEK, Moss SE: Visual im-

pairment in diabetes. Ophthalmology 91:

1–9, 1984

2. Leslie RD, Pyke DA: Diabetic retinopathy

in identical twins. Diabetes 31:19–21, 1982

3. Fernandez LA, Twickler J, Mead A: Neo-

vascularisation produced by angiotensin

II. J Lab Clin Med 105:141–145, 1985

4. Danser AH, Derkx FH, Admiraal PJ, Dei-

num J, de Jong PT, Schalekamp MA:

Angiotensin levels in the eye. Invest Oph-

thalmol Vis Sci 35:1008–1018, 1994

5. Migdalis IN, Iliopoulou V, Kalogeropou-

lou K, Koutoulidis K, Samartzis M:

Elevated serum levels of angiotensin-con-

verting enzyme in patients with diabetic

retinopathy. South Med J 83:425–427, 1990

6. Chaturvedi N, Sjolie A-K, Stephenson JM,

Abrahamian H, Keipes M, Castellarin A,

Rogulja-Pepeonik Z, Fuller J, the EUCLID

Study Group: Effect of lisinopril on pro-

gression of retinopathy in normotensive

people with type 1 diabetes. Lancet 351:

28–31, 1998

7. Rabensteiner D, Abrahamian H, Irsigler

K, Hermann KM, Kiener HP, Mayer G,

Kaider A, Prager R: ACE gene polymor-

phism and proliferative retinopathy in

type 1 diabetes: results of a case-control

study. Diabetes Care 22:1530–1535, 1999

8. Fujisawa T, Ikegami H, Kawaguchi Y,

Hamada Y, Ueda H, Shintani M, Fukuda

M, Ogihara T: Meta-analysis of associa-

tion of insertion/deletion polymorphism

of angiotensin I-converting enzyme gene

with diabetic nephropathy and retinopa-

thy. Diabetologia 41:47–53, 1998

9. Tarnow L, Cambien F, Rossing P, Nielsen

FS, Hansen BV, Ricard S, Poirer O, Parv-

ing HH: Angiotensin II type 1 receptor

gene polymorphism and diabetic micro-

angiopathy. Nephrol Dial Transplant 11:

1019–1023, 1996

10. Staessen JA, Kuznetsova T, Wang JG,

Emelianov D, Vlietinck R, Fagard R:

M235T angiotensinogen gene polymor-

phism and cardiovascular renal risk. J Hy-

pertens 17:9–17, 1999

11. Ko GTC, Chan JCN, Lau MSW, Cockram

CS: Diabetic microangiopathic complica-

tions in young Chinese diabetic patients: a

clinic-based cross-sectional study. J Dia-

betes Complications 13:300–306, 1999

12. Thomas GN, Critchley JAJH, Tomlinson

B, Lee ZSK, Young RP, Chan JCN: Al-

buminuria and the renin-angiotensin

system gene polymorphisms in type 2

diabetic and in normoglycaemic hyper-

tensive Chinese. Clin Nephrol 55:7–15,

2001

13. Peters J: Molecular basis of human hyper-

tension: the role of angiotensin. Baillieres

Clin Endocrinol Metab 9:657–678, 1995

Prevalence of Type 1

Diabetes–Related

Autoantibodies in

Adults With Celiac

Disease

tudies have shown a 4% prevalence

of celiac disease in type 1 diabetic

patients. Few data, however, are

available on the prevalence of type 1 dia-

betes–related antibodies in patients with

celiac disease, with studies being limited

by the recruitment of low numbers of

children only (1,2). We have assessed

prevalence of type 1 diabetes–related au-

toantibodies (islet cell antibody [ICA],

GAD antibody [GADA], and antibodies to

the protein tyrosine phosphatase–related

IA-2 [anti-IA2]) in a cohort of 378 nondi-

abetic adults with celiac disease (89 un-

treated and 289 treated with a gluten-free

diet), aged 33.9 ⫾14.7 years (range

15.7–81.3), who were cared for at our

outpatient clinic. Known duration of glu-

Table 1—Genotype distributions for the ACE I/D, AGT M235T, and angiotensin type 1

receptor A1166C gene polymorphisms in type 2 diabetic patients without and with retinopathy

Genotypes

retinopathy

Retinopathy

(total) Nonproliferative

Advanced retinopathy

(pre, proliferative,

advanced)

n501 326 225 101

ACE DD/ID/II 11.6/42.3/46.1 12.3/39.6/48.2 11.4/37.8/50.7 14.9/42.6/42.6

AGT TT/MT/MM 69.7/28.5/1.7 74.2/24.0/1.8 75.0/23.8/1.2 77.0/20.7/2.3

R AC/AA 9.2/90.8 6.6/93.4 8.1/91.9 6.5/93.5

No signiﬁcant differences were identiﬁed between the groups.

Letters

1644 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

ten withdrawal was ⬍5 years in 146

(48.2%), 5–9 years in 54 (17.8%), and

⬎9 years in 103 (34.0%) patients. GADAs

were measured by a radioligand assay us-

ing human recombinant GAD 65 as anti-

gen (Medipan Diagnostica, Selchow,

Germany), ICAs by indirect immunoﬂu-

orescence on frozen sections of human

blood group 0 pancreas with ﬂuorescein

isothiocyanate-conjugated rabbit anti-

bodies, and anti-IA2 by a radioligand as-

say using highly puriﬁed human

recombinant IA2 labeled with

125

I (Medi-

pan Diagnostica).

Of 289 treated nondiabetic patients,

26 had type 1 diabetes–related autoanti-

bodies, giving a prevalence of 9.0% (95%

CI 6.7–12.3). All of them, however,

showed no more than one marker positiv-

ity (2.8% ICA, 3.1% GADA, 3.1% anti-

IA2). Untreated celiac patients had a

similar prevalence (10.0%, P⫽0.75). In

linear regression analysis, levels of anti-

IA2 were linearly associated with dura-

tion of celiac disease (␤⫽0.002, P⫽

0.05). In logistic regression analysis, du-

ration of gluten withdrawal was indepen-

dently associated with a prevalence of

type 1 diabetes–related autoantibodies.

With respect to duration ⬍5 years, a four-

fold increased risk in patients with dura-

tion ⬎9 years was found (95% CI 1.51–

10.6), even after adjustment for age, sex,

presence of other autoimmune diseases,

and compliance to diet. In a mean fol-

low-up time of 3.0 ⫾1.3 years (range

1.3–6.5), however, no incident case of

type 1 diabetes was diagnosed in either

patients with (79.0 person-years) or with-

out (782.7 person-years) autoimmune

markers.

In conclusion, this study shows that

prevalence of type 1 diabetes–related au-

toantibodies in adults with celiac disease

is high (9.0%), even after dietary gluten

withdrawal, and that it increases over

time but is associated with low risk of pro-

gression to diabetes. These ﬁndings are

consistent with a role of common genetic

susceptibility of both diseases, such as

factors involved in intestinal permeabil-

ity. Case-control studies show association

between various dietary factors and risk of

type 1 diabetes (3). Dietary gluten could

act as modiﬁer rather than determinant of

type 1 diabetes, facilitating the progres-

sion of other dietary factors to the lamina

propria, where they activate the autoim-

mune response against ␤-cells.

GRAZIELLA BRUNO,

SILVIA PINACH,

SILVIA MARTINI,

MAURIZIO CASSADER,

GIANFRANCO PAGANO,

CARLA SATEGNA GUIDETTI,

From the Department of Internal Medicine, Univer-

sity of Torino, Torino, Italy.

Address correspondence to Dr. Graziella Bruno,

Department of Internal Medicine, Turin University,

corso Dogliotti 14, I-10126 Torino, Italy. E-mail:

graziella.bruno@katamail.com.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Di Mario U, Anastasi E, Mariani P, Ballati

G, Perfetti R, Triglione P, Morellini M,

Bonamico M: Diabetes-related autoanti-

bodies do appear in children with coeliac

disease. Acta Paediatr 81:593–597, 1992

2. Galli-Tsinopoulou A, Nousia-Arvanitakis

S, Dracoulacos D, Xefteri M, Karamouzis

M: Autoantibodies predicting diabetes

mellitus type 1 in celiac disease. Horm Res

52:119–124, 1999

3. Dahlquist GG, Blom LG, Persson LA, Sad-

strom AI, Wall SG: Dietary factors and the

risk of developing insulin dependent dia-

betes in childhood. BMJ 300:1302–1306,

1990

Glycemic Relapse

After an Intensive

Outpatient

Intervention for Type

2 Diabetes

he effectiveness of interventions that

reduce hyperglycemia in patients

with diabetes (1,2) is limited by the

tendency for glycemic relapse after the in-

tervention ends. We sought to character-

ize the occurrence of glycemic relapse

after initial improvement and to describe

predictors of relapse.

Glycemic relapse was evaluated in

265 consecutive patients with type 2 dia-

betes who participated in a 3-month in-

tensive outpatient intervention and were

followed at least 1 year after completion of

the intervention. Details of the interven-

tion have been previously described (3).

All had HbA

⬎8% before the interven-

tion and had achieved marginal glycemic

control (nadir HbA

⬍8%) after the in-

tervention. Glycemic relapse was deﬁned

as a subsequent HbA

ⱖ8% and an ab-

solute increase of at least 1% above the

postintervention nadir HbA

. All data

were analyzed using SAS 8.12 (SAS Insti-

tute, Cary, NC).

The mean ⫾SD age was 56 ⫾13.4

years; 54% were women; 71% were Cau-

casian; and 27% were African American.

The mean duration of diabetes was 6.4 ⫾

8.4 years. Mean BMI was 32.2 ⫾8.6 kg/

. The mean HbA

before interven-

tion was 10.1 ⫾1.7%. The mean HbA

nadir after intervention was 6.8 ⫾0.7%.

Twenty-ﬁve percent were receiving insu-

lin therapy before the intervention, and

an additional 25% initiated insulin during

the program.

The cumulative incidence of relapse

at 1 year was 25%. The initiation of insu-

lin therapy during the intervention was

the only identiﬁable independent predic-

tor of relapse (hazard ratio 1.96, 95% CI

1.02–3.74). Female sex, African-

American race, duration of diabetes, lack

of weight loss during the intervention pe-

riod, and the levels of HbA

before the

intervention and at nadir after the inter-

vention increased the risk of relapse in

univariate analysis, but these associations

were not statistically signiﬁcant in multi-

variate modeling. For those patients who

relapsed, the median time to relapse was 9

months. Kaplan-Meier plots estimated

that 50% of the population would relapse

by 30.3 months.

Twenty-ﬁve percent of type 2 patients

who attained satisfactory glycemic con-

trol after an intensive outpatient interven-

tion relapsed within a year. Patients who

initiated insulin therapy during the inter-

vention had an almost double risk of re-

lapse. Future studies should characterize

social and behavioral variables as well as

the frequency and type of insulin in order

to better understand the relapse process.

In the interim, this high-risk subgroup

should receive priority for continuation of

intensive care or other relapse preventive

measures.

Acknowledgments—This study was sup-

ported in part by NIDDK P60DK20593, the

American Diabetes Association, and the Rob-

ert Wood Johnson Foundation.

TOM A. ELASY,

MD, MPH

ALAN L. GRABER,

KATHLEEN WOLFF,

RN, MSN

ANNE BROWN,

RN, MSN

AYUMI SHINTANI,

PHD

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1645

From the

Department of Medicine, Vanderbilt Uni-

versity School of Medicine, Nashville, Tennessee;

and

Vanderbilt University School of Nursing,

Nashville, Tennessee.

Address correspondence to Tom A. Elasy, S1121

MCN, Department of Medicine, Vanderbilt Univer-

sity School of Medicine, Nashville, TN 37232-2587.

E-mail: tom.elasy@mcmail.vanderbilt.edu.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. The Diabetes Control and Complications

Trial Research Group: The effect of inten-

sive treatment of diabetes on the develop-

ment and progression of long-term

complications in insulin-dependent dia-

betes mellitus. N Engl J Med 329:977–986,

1993

2. UK Prospective Diabetes Study (UKPDS)

Group: Intensive blood-glucose control

with sulphonylureas or insulin compared

with conventional treatment and risk of

complications in patients with type 2 di-

abetes (UKPDS 33). Lancet 352:837–853,

1998

3. Graber AL, Elasy TA, Quinn D, Wolff K,

Brown A: Improving glycemic control in

adults with diabetes mellitus: shared re-

sponsibility in primary care practices.

South Med J 95:684–690, 2002

Gestational Glucose

Intolerance in

Multiple Pregnancy

win pregnancies have been associ-

ated with adverse obstetric and neo-

natal outcomes. The incidence of

respiratory distress syndrome in the new-

borns of multiple pregnancy patients is

probably related to the higher incidence

of preterm delivery. However, the com-

plications of twin pregnancies are not

limited to infants; miscarriages, pre-

eclampsia, anemia, placenta previa, poly-

hydramnios, and premature rupture of

membranes are more frequent than in sin-

gleton pregnancy. There have been a few

reports that test the hypothesis that mul-

tiple gestation could also be a risk factor

for gestational diabetes (1–6). These

studies utilized various criteria for diag-

nosing gestational diabetes (World

Health Organization [1], National Diabe-

tes Data Group [2–4], and Carpenter and

Coustan [7]) and did not report the per-

centage of the positive glucose challenge

test for each group.

To determine the prevalence of gesta-

tional diabetes (using the Carpenter and

Coustan criteria) in singleton and twin

pregnancies in a cohort population of

women who had had screening for ges-

tational diabetes from January 1990 to

January 2000, we started a retrospec-

tive study on 2,554 pregnant women

screened for gestational diabetes in the

Department of Obstetrics and Gynecol-

ogy of our University Hospital. Of these,

70 were multiple pregnancies. A 50-g oral

glucose challenge test was performed be-

tween 24 and 28 weeks’gestation, and a

1-h value ⱖ135 mg/dl was considered

positive and followed by a 100-g oral glu-

cose tolerance test. The results were inter-

preted according to the Carpenter and

Coustan threshold values (95, 180, 155,

and 140 mg/dl) at fasting and after 1, 2,

and 3 h, respectively. If two thresholds

were met or exceeded, then the diagnosis

of gestational diabetes was made. Age, fa-

milial or personal anamnestic factors, par-

ity, and BMI at ﬁrst visit, all considered

risk factors for GDM, were recorded in

our database. A positive glucose challenge

test occurred in 520 of 2,554 pregnant

women (20.3%). The test was positive in

22 of 70 (31.4%) multiple pregnancies

and 498 of 2,484 (20.0%) singleton preg-

nancies (P⫽0.029). The rates of GDM

were 80 of 2,484 (3.6%) in singleton and

4 of 70 (5.7%) in twin pregnancies (P⫽

0.416). No statistically signiﬁcant differ-

ences exist for the maternal risk factors for

GDM between the two groups.

The decreased insulin sensitivity in

pregnancy may be modiﬁed by several

factors, such as diet, BMI, maternal age,

and the placental mass, all of which may

play a role affecting ␤-cell function and

sensitivity to insulin. It has been sug-

gested that in multiple pregnancies with

two placentas or one that is larger, the

incidence of gestational diabetes may be

increased (3,4,6). Our data show that in

our Sicilian population, twin pregnancy

cannot be considered a proven risk factor

for gestational diabetes. In fact, the differ-

ences (3.6 vs. 5.7%) are not statistically

different. Our data showed that one-third

of twin pregnancies (22 of 70, 31.4%)

manifest a positive glucose challenge test,

as compared with 20% of singleton preg-

nancies (498 of 2,484). It seems possible

that with a larger sample size, the differ-

ence in incidence of gestational diabetes

might have been statistically signiﬁcant,

although of questionable clinical signiﬁ-

cance. Another hypothesis could be that

the greater relative elevation of anti-

insulin placental hormone levels in a twin

gestation may precipitate a mild degree of

glucose intolerance in a woman who

would have been normal with a singleton

pregnancy. If this were the case, the im-

pact of such a mild degree of glucose in-

tolerance remains undetermined.

Acknowledgments—The authors thank

Prof. D.R. Coustan (Brown University, Provi-

dence, RI) for critical reading of the manu-

script.

FRANCESCO CORRADO,

FRANCESCO CAPUTO,

GRAZIELLA FACCIOLA` ,

ALFREDO MANCUSO,

From the Department of Obstetrics and Gynecol-

ogy, University of Messina, Messina, Italy.

Address correspondence to F. Corrado, MD, De-

partment of Obstetrics and Gynecology, Policlinico

Universitario “G. Martino”98100, Messina, Italy. E-

mail: francesco.corrado@unime.it.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Wein P, Warwick MM, Beisher NA: Ges-

tational diabetes in twin pregnancy: prev-

alence and long-term implications. Aust

NZ J Obstet Gynecol 32:325–327, 1992

2. Henderson CE, Scarpelli S, La Rosa D, Di-

von MY: Assessing the risk of gestational

diabetes in twin gestation. J Natl Med As-

soc 87:757–758, 1995

3. Roach VJ, Lau TK, Wilson D, Rogers MS:

The incidence of gestational diabetes in

multiple pregnancy. Aust NZ J Obstet Gy-

necol 38:56–57, 1998

4. Schwartz DB, Daoud Y, Zazula P, Goyert

G, Bronsteen R, Wright D, Copes J: Ges-

tational diabetes mellitus: metabolic and

blood glucose parameters in singleton

versus twin pregnancies. Am J Obstet Gy-

necol 181:912–914, 1999

5. Egeland GM, Irgens LM: Is multiple birth

pregnancy a risk factor for gestational di-

abetes? (Letter). Am J Obstet Gynecol 185:

1275–1276, 2001

6. Sivan E, Maman E, Homko CJ, Lipitz S,

Cohen S, Schiff E: Impact of fetal reduc-

tion on the incidence of gestational diabe-

tes. Obstet Gynecol 99:91–94, 2002

7. Carpenter MW, Coustan DR: Criteria for

screening test for gestational diabetes.

Am J Obstet Gynecol 144:768–773, 1982

Letters

1646 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

Cytokine Milieu

Tends Toward

Inﬂammation in

Type 2 Diabetes

t has been postulated that type 2 dia-

betes is a manifestation of the inﬂam-

matory host response (1). Increased

inﬂammatory activity is believed to play

a critical role in development of athero-

genesis and to predispose established

atherosclerotic plaques to rupture (2). In-

terleukin (IL)-8 is a potent chemoattractant

(chemokine) and induces recruitment of

neutrophils and T-cells into the subendo-

thelial space, as well as adhesion of mono-

cytes to endothelium (3). IL-18 is a potent

proinﬂammatory cytokine reported to

play a role in plaque destabilization and to

predict cardiovascular death in patients

with coronary artery disease (4). Adi-

ponectin is an adipocyte-derived plasma

protein (adipokine) that accumulates in

the injured artery and has potential anti-

atherogenic properties; male patients

with hypoadiponectinemia present a two-

fold increase in coronary artery disease

prevalence (5).

We studied 30 patients (age 42 ⫾4

years, BMI 26.9 ⫾1.2 kg/m

, and HbA

7.2 ⫾0.6% [mean ⫾SD]) with newly

diagnosed type 2 diabetes (within 6

months of diagnosis) without hyperten-

sion or evidence of vascular complica-

tions treated only by diet. Thirty healthy

subjects, matched for age, sex, and body

weight, served as the control group. None

of the subjects (diabetic and nondiabetic)

were taking any drugs. Serum samples for

IL-8, IL-18, and adiponectin were stored

at ⫺80°C and measured in duplicate us-

ing enzyme-linked immunosorbent kits.

All samples for a given patient were ana-

lyzed in the same series.

Circulating levels of IL-8 and -18 con-

centrations were higher in diabetic pa-

tients (13.7 ⫾2.8 and 205 ⫾39 pg/ml,

respectively) than in control subjects

(8.7 ⫾1.9 and 120 ⫾25 pg/ml, P⬍

0.05–0.001), while circulating adiponec-

tin levels were lower (4.9 ⫾1.2 vs. 7.1 ⫾

1.9 ␮g/ml, P⬍0.01). There was a signif-

icant correlation between fasting glucose

and IL-8 levels in diabetic patients (r⫽

0.31, P⬍0.05); adiponectin concentra-

tions were negatively correlated with fast-

ing insulin levels (r⫽⫺0.43, P⬍0.01).

Our study shows that circulating IL-8

concentrations are signiﬁcantly higher

and adiponectin levels are signiﬁcantly

lower in type 2 diabetic patients than

matched control subjects. Previous stud-

ies reported similar ﬁndings (6,7). To our

knowledge, this is the ﬁrst demonstration

that circulating IL-18 concentrations are

increased in type 2 diabetic patients, as

compared with age-, sex-, and body

weight–matched nondiabetic subjects. In

a prospective study of 1,229 subjects, in-

cluding one-sixth of diabetic patients,

with documented coronary artery dis-

ease, serum IL-18 concentration was

identiﬁed as a strong independent predic-

tor of future cardiovascular events (5).

The correlation we found between IL-18

and glucose levels in diabetic patients and

the evidence that acute hyperglycemia

may increase circulating IL-18 levels in

healthy subjects and IGT patients (8) sug-

gest a role for this cytokine in plaque de-

stabilization associated with stress

hyperglycemia (9).

KATHERINE ESPOSITO,

FRANCESCO NAPPO,

MD, PHD

FRANCESCO GIUGLIANO

CARMEN DIPALO

MYRIAM CIOTOLA

MICHELANGELA BARBIERI,

GIUSEPPE PAOLISSO,

DARIO GIUGLIANO,

MD, PHD

From the Department of Geriatrics and Metabolic

Diseases, Second University of Naples, Naples, Italy.

Address correspondence to Katherine Esposito,

MD, Dipartimento di Geriatria e Malattie del Me-

tabolismo, Policlinico Universitario, Piazza L. Mira-

glia,80138Napoli,Italy.E-mail: katherine.esposito@

unina2.it.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Pickup JC, Mattock MB, Chusney GD,

Burt D: NIDDM as a disease of the in-

nate immune system: association of

acute-phase reactants and interleukin-6

with metabolic syndrome. Diabetologia 40:

1286–1292, 1997

2. Libby P, Ridker PM, Maseri A: Inﬂamma-

tion and atherosclerosis. Circulation 105:

1135–1143, 2002

3. Gerzten RE, Garcia-Zepeda EA, Limm YC,

Yoshida M, Ding HA, Gimbrone Jr MA,

Luster AD, Luscinskas FW, Rosenzweig

A: MCP-1 and IL-8 trigger ﬁrm adhesion

of monocytes to vascular endothelium

under ﬂow conditions. Nature 398:718–

723, 1999

4. Blankenberg S, Tiret L, Bickel C, Peetz D,

Cambien F, Meyer J, Rupprecth HJ: Inter-

leukin-18 is a strong predictor of cardio-

vascular death in stable and unstable

angina. Circulation 106:24–30, 2002

5. Kumada M, Kihara S, Sumitsuji S,

Kawamoto T, Matsumoto S, Ouchi N,

Arita Y, Okamoto Y, Shimomura I,

Hiraoka H, Nakamura T, Funahashi T,

Matsuzawa Y, the Osaka CAD Study

Group: Association of hypoadiponectine-

mia with coronary artery disease in man.

Arterioscler Thromb Vasc Biol 23:85–89, 2003

6. Zozulin˜ska D, Majchrzak A, Sobieska M,

Wiktorowicz K, Wierusz-Wysocka B:

Serum interleukin-8 levels is increased

in diabetic patients. Diabetologia 42:117–

123, 1999

7. Weyer C, Funahashi T, Tanaka S, Hotta K,

Matsuzawa Y, Pratley RE, Tataranni PA:

Hypoadiponectinemia in obesity and type

2 diabetes: close association with insulin

resistance and hyperinsulinemia. J Clin

Endocrinol Metab 86:1930–1935, 2001

8. Esposito K, Nappo F, Marfella R, Giugli-

ano G, Giugliano F, Ciotola M, Quagliaro

L, Ceriello A, Giugliano D: Inﬂammatory

cytokine concentrations are acutely in-

creased by hyperglycemia in humans: role

of oxidative stress. Circulation 106:2067–

2072, 2002

9. Capes SE, Hunt D, Malberg K, Gerstein

HC: Stress hyperglycemia and increased

risk of death after myocardial infarction in

patients with and without diabetes: a sys-

temic overview. Lancet 355:773–778, 2000

Lipoprotein

Abnormalities in

Human Genetic

CD36 Deﬁciency

Associated With

Insulin Resistance

and Abnormal Fatty

Acid Metabolism

D36 is an 88-kDa membranous pro-

tein with multiple relevant function

that is widely expressed in human

tissues (1). Because we and others (2)

found human genetic deﬁciency of this

molecule, we have elucidated the molec-

ular bases and pathophysiology. We have

reported that the human deﬁciency may

be associated with insulin resistance (3)

and abnormal dynamics of long-chain

fatty acid (LCFA) (4). The aim of the

present study was to further characterize

lipid and lipoprotein metabolism in the

human genetic CD36 deﬁciency, espe-

cially focusing on postprandial responses.

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1647

Seven Japanese patients with type I

CD36 deﬁciency (cases 1–7) were sub-

jected. They were divided into the follow-

ing two groups: group A, two younger

patients we have recently identiﬁed (cases

1 and 2); and group B, patients ⬎50 years

of age (cases 3–7, reported in our previ-

ous paper [3]). For the analysis of each

group, aged-matched healthy and nono-

bese volunteers were subjected as control

subjects (Table 1). First, we conﬁrmed

that insulin resistance was also present in

the two younger patients (group A).

At the fasting state, “midbands,”

which imply the accumulation of rem-

nant lipoproteins, were observed in group

B but not in group A. Preparative ultra-

centrifugation conﬁrmed an increase in

intermediate-density lipoprotein choles-

terol in the all of the CD36-deﬁcient sub-

jects in group B compared with the

control subjects. The oral fat loading test

demonstrated that the patients in group B

presented postprandial hyperlipidemia

with the accumulation of small intestine–

derived lipids. Whereas the patients in

group A did not show apparent abnor-

malities, though they did have insulin

resistance.

The present study indicated that this

monogenic disorder with the abnormal

metabolism in LCFA and insulin resis-

tance may cause the phenotypic expres-

sion of hyperlipidemia at both fasting and

postprandial states. These data also sug-

gest that the clinical manifestation is af-

fected by the acquired factors, such as

aging and dietary contents. We believe

that human genetic CD36 deﬁciency may

be a monogenic form of “metabolic syn-

drome.”Further studies to elucidate mo-

lecular mechanism for the phenotypic

expression in CD36-deﬁcient states are

under way in our laboratories.

Acknowledgments—This work was sup-

ported by grants-in-aid to K.H. (no.

13671191) and S.Y. (nos. 11557055 and

10671070) from the Ministry of Education,

Science, Sports, and Culture of Japan.

We gratefully thank Natsu Iwamoto, Sonoe

Shima, and Atsuko Ohya for the skillful tech-

nical assistance.

TAKAHIRO KUWASAKO,

KEN-ICHI HIRANO,

MD, PHD

NAOHIKO SAKAI,

MD, PHD

MASATO ISHIGAMI,

MD, PHD

HISATOYO HIRAOKA,

MD, PHD

MOHAMED JANABI YAKUB,

MD, PHD

KEIKO YAMAUCHI-TAKIHARA,

MD, PHD

SHIZUYA YAMASHITA,

MD, PHD

YUJI MATSUZAWA,

MD, PHD

From the

Department of Internal Medicine and

Molecular Science, Graduate School of Medicine,

Osaka University, Osaka, Japan; and the

Depart-

ment of Molecular Medicine, Graduate School of

Medicine, Osaka University, Osaka, Japan.

Address correspondence to Ken-ichi Hirano,

MD, PhD, Department of Internal Medicine and Mo-

lecular Science, Graduate School of Medicine, B5,

Osaka University, 2-2 Yamadaoka, Suita, Osaka

565-0871, Japan. E-mail: khirano@kb3.so-net.ne.jp.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Febbraio M, Hajjar DP, Silverstein RL:

CD36: a class B scavenger receptor in-

volved in angiogenesis, atherosclerosis,

inﬂammation, and lipid metabolism.

J Clin Invest 108:785–791, 2001

2. Yamamoto N, Ikeda H, Tandon NN, Her-

man J, Tomiyama Y, Mitani T, Sekiguchi

S, Lipsky R, Kralisz U, Jamieson GA: A

platelet membrane glycoprotein (GP) de-

ﬁciency in healthy blood donors: Nak

platelet lack detectable GP IV(CD36).

Blood 76:1698–1703, 1990

3. Miyaoka K, Kuwasako T, Hirano K,

Nozaki S, Yamashita S, Matsuzawa Y:

CD36 deﬁciency associated with insulin

resistance. Lancet 357:686–687, 2001

4. Yoshizumi T, Nozaki S, Fukuchi K, Ya-

masaki K, Fukuchi T, Maruyama T, To-

miyama Y, Yamashita S, Nishimura T,

Matsuzawa Y: Pharmacokinetics and

metabolism of iodine-123-BMIPP fatty

acid analogue in normal and CD36-deﬁ-

cient subjects. J Nucl Med 41:1134–1138,

2000

The Relation

Between Physical

Activity and

Metabolic Control in

Type 2 Diabetes

With <20 Years of

Evolution

besity and reduced levels of physi-

cal activity are directly associated

with the insulin resistance that

characterizes type 2 diabetes (1,2). Phys-

ical activity is one of the cornerstones of

type 2 diabetes management (3). Few

Table 1—Lipid and lipoprotein abnormalities at fasting and postprandial states in human genetic CD36 deﬁciency

Group A Group B

Case 1 Case 2 Control (n⫽4) Case 3 Case 4 Case 5 Case 6 Case 7 Control (n⫽8)

Age (years)/sex (M/F) 34/M 32/M 33 ⫾2 (4M) 58/M 65/F 54/F 54/F 54/M 55 ⫾10 (3M/5F)

BMI (kg/m

) 22.9 32.0 22.4 ⫾1.0 26.2 22.2 32.8 17.3 24.9 22.7 ⫾2.3

WBGU (mg 䡠kg

⫺1

䡠min

⫺1

)* 7.8 2.8 9.1 ⫾0.5 2.67†7.09†5.49†4.85†5.29†8.6 ⫾0.5†

Fasting state

Total cholesterol (mmol/l) 4.20 5.01 4.78 ⫾0.38 5.81 6.18 5.97 5.45 5.84 5.10 ⫾0.69

Triglyceride (mmol/l) 0.78 1.23 1.00 ⫾0.17 2.73 1.43 1.77 1.28 3.41 1.20 ⫾0.38

Midband in PAG disc

electrophoresis‡

⫺⫺ ⫹ ⫹ ⫹⫹⫹

IDL cholelsterol (mmol/l)§ND ND 0.48 0.38 0.29 0.34 0.22 0.11 ⫾0.03

Postprandial state㛳

AUC-TG (mmol/l 6 h) 5.3 8.8 8.8 ⫾0.7 20.6 13.4 14.0 14.0 26.8 10.0 ⫾2.9

AUC-Apo B48 (AU/ml 6 h)** 223 268 329 ⫾107 866 671 941 551 541 360 ⫾120

Data are means ⫾SD unless otherwise indicated. apo, apolipoprotein; AUC, area under the curve; PAG, polyacrylamide gel; WBGU, mean whole-body glucose

uptake. *Insulin resistance was determined by euglycemic clamp technique (ref. 3); †part of data from cases 3–7 were described previously (Ref. 3); ‡lipoproteins

were analyzed by PAG electrophotesis. §IDL was separated by preparative ultracentrifugation (1.019 ⬍d⬍1.063 g/ml); 㛳postprandial hyperlipidemia was analyzed

by the oral fat loading test with fat cream containing 30 g fat/m

body surface area; TG and apo B48 levels were measured at 0, 2, 4, and 6 h after loading, and AUCs

were calculated; **serum apo B48 levels were measured by the enzyme-linked immunosorbent assay we have developed.

Letters

1648 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

large studies have assessed the impact of

intensity and quality of exercise on meta-

bolic control in these patients, and the

inﬂuence of age on the metabolic re-

sponse to physical activity is still contro-

versial (4). We conducted the Intervention

on Key Cardiovascular Risk Factors in

Type 2 Diabetes National Study (KAREN),

a country-based case-control study nested

in a survey of 1,485 consecutive type 2

diabetic patients treated by 208 endocri-

nologists across Spain to assess the impact

of physical activity on metabolic control.

Patient risk factors, history characteris-

tics, medication and diet compliance, gly-

cemia, fasting HbA

, and lipids were

determined.

Patient characteristics are presented

in Table 1 by quartile of physical activity.

After adjustment for age, sex, triglycer-

ides, BMI, and diet compliance, patients

in the lowest quartile of physical activity

had an OR of 1.52 and a 95% CI of 1.11–

2.08 for having an HbA

⬎7.5%. The

ORs for patients ⬍65 years of age and in

the lowest quartile of physical activity

were 1.55 (95% CI 1.04 –2.31) compared

with the fourth, 1.64 (1.09–2.47) with

the third, and 1.84 (1.22–2.75) with the

second quartile. Patients ⬎65 years of age

and in the lowest quartile of physical ac-

tivity had a similar adjusted risk com-

pared with the rest of quartiles (OR 1.14,

95% CI 0.70–1.85; 1.38, 0.84–2.26, and

0.87, 0.53–1.43, respectively). A signiﬁ-

cant reduction of 0.2% (95% CI ⫺0.36 to

⫺0.04) in HbA

was observed for pa-

tients expending ⬎50 kcal/day versus the

ﬁrst quartile; the reduction for patients

ⱕ65 years of age was 0.27% (⫺0.49 to

–0.06), but for patients ⬎65 there was no

signiﬁcant difference.

These results suggest that physical ac-

tivity produces better glycemic control

and improved cardiovascular risk proﬁle

in diabetic patients, particularly in those

⬍65 years of age.

PEDRO DE PABLOS-VELASCO,

MD, PHD

WILFREDO RICART,

MD, PHD

SUSANA MONEREO,

BASILIO MORENO,

MD, PHD

JAUME MARRUGAT,

MD, PHD, FESC

ON BEHALF OF THE KAREN INVESTIGATORS

From the

Department of Endocrinology, Hospital

Doctor Negrı´n, Las Palmas de Gran Canaria, Spain;

the

Department of Endocrinology, Hospital Trueta,

Gerona, Spain; the

Department of Endocrinology,

Hospital Getafe, Madrid, Spain; the

Department of

Endocrinology, Hospital Gregorio Maran˜on, Ma-

drid, Spain; and the

Unit of Lipid and Cardiovas-

cular Epidemiology, Institut Municipal d’Investigacio´

Me`dica (IMIM) and Universitat Auto´noma School

of Medicine, Barcelona, Spain.

Address correspondence to Dr. Jaume Marrugat,

Lipid and Cardiovascular Epidemiology Unit, Insti-

tut Municipal d’Investigacio´Me`dica (IMIM), Carrer

Dr. Aiguader, 80, E-08003 Barcelona, Spain. E-mail:

jmarrugat@imim.es.

Appendix

The KAREN Investigators comprise 208

Spanish endocrinologists who work in

primary outpatient diabetes clinics of the

National Health Service.

The KAREN investigators are as fol-

lows: Abella´n T, Acha FJ, Acosta D, Agui-

lar M, Aguillo E, Albero R, Alcaraz M,

Alvarez JA, Alvarez P, Alvarez J, Andia

VM, Aranda S, Arranz A, Arribas LD, Ar-

teaga R, Asensio C, Ayala C, Ballester E,

Barril R, Becerra A, Blanco B, Blay V, Botas

P, Brito MA, Bruscas M, Bueno J, Burgo R,

Cabezas J, Caldero´n D, Calero M, Calvo F,

Ca´ncer E, Carrasco R, Casal F, Castan˜oG,

Castro JC, Cepero D, Cervello G, Chena

JA, Conget I, Corrales JJ, Costilla E, Coves

MJ, Darı´as R, de la Cuesta C, de Pablos P,

Dı´az C, Diez A, Domı´nguez JR, Duran S,

Duran M, Elviro R, Enrı´quez L, Escalada

FJ, Escobar L, Esmatges E, Espiga J, Es-

topin˜an V, Ezquerra R, Famades A, Fer-

na´ndez JI, Ferna´ndez M, Ferna´ndez JM,

Ferna´ndez F, Ferna´ndez T, Ferna´ndez JA,

Ferna´ndez J, Fla´ndez B, Flaquert P, Fraile

AL, Fraile J, Freijanes J, Galvez A, Garcı´a

Table 1—Demographic, comorbidity, lifestyle, biochemical, and somatometric characteristics by physical activity levels (quartiles) of type 2

diabetic patients who participated in the KAREN study

⬍51 kcal/day 51–148 kcal/day 149–264 kcal/day ⬎264 kcal/day P

n368 373 375 369

Age (years) 62 ⫾10 61 ⫾10 61 ⫾961⫾10 NS

Women (%) 49.6 45.3 54.4 69.6 ⬍0.001

Time with diabetes (years) 10.3 ⫾6.4 9.9 ⫾6.3 10.1 ⫾6.4 10.3 ⫾6.2 NS

Hypertension (%) 51.8 58.3 50.7 46.3 0.017

Current smokers (%)* 19.4 18.9 19.0 16.0 NS

Systolic BP (mmHg) 140 ⫾17 139 ⫾17 139 ⫾16 139 ⫾19 NS

Diastolic BP (mmHg) 80 ⫾10 79 ⫾979⫾10 78 ⫾11 NS

BMI (kg/m

) 28.8 ⫾3.4 28.5 ⫾3.5 28.1 ⫾3.3 28.3 ⫾3.3 0.035

Coronary heart disease (%)†17.9 14.4 18.0 14.0 NS

Cerebrovascular disease (%)‡8.5 6.4 6.6 5.1 NS

Glycemia (mg/dl) 9.5 ⫾3.2 9.1 ⫾2.9 9.1 ⫾3.0 8.8 ⫾2.7 ⬍0.001

HbA

(%)* 7.9 ⫾1.6 7.6 ⫾1.4 7.6 ⫾1.4 7.4 ⫾1.0 ⬍0.001

Total cholesterol (mmol/l) 5.3 ⫾1.0 5.4 ⫾1.0 5.2 ⫾1.0 5.1 ⫾1.0 ⬍0.001

HDL cholesterol (mmol/l) 1.2 ⫾0.3 1.3 ⫾0.3 1.2 ⫾0.3 1.3 ⫾0.4 0.014

LDL cholesterol (mmol/l) 3.4 ⫾0.9 3.4 ⫾0.9 3.2 ⫾34.7 3.3 ⫾0.9 NS

Triglycerides (mmol/l) 1.9 ⫾1.8 1.7 ⫾0.9 1.7 ⫾1.1 1.5 ⫾0.9 ⬍0.001

Creatinine (␮mol/l) 91.9 ⫾31.8 85.7 ⫾24.7 86.6 ⫾26.5 88.4 ⫾19.4 0.035

Good diet compliance (%) 37.1 35.1 49.2 52.2 ⬍0.001

Daily alcohol consumption (g) 12.9 ⫾12.8 12.5 ⫾12.3 12.8 ⫾13.3 12.5 ⫾12.3 NS

Data are means ⫾SD and %. *Smokers of at least one cigarette per day in average; †myocardial infarction or angina; ‡transient ischemic attack or stroke.

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1649

JA, Garcı´a A, Garcı´a C, Garcı´a G, Garcı´aI,

Garcı´a J, Garcı´a M, Garcia-Hierro V, Gar-

gallo MA, Gentil AC, Gil A, Gilsanz A,

Gime´nez G, Gippini A, Girbe´s AJ, Goday

A, Godoy DV, Goena M, Goicolea I, Go´-

mez F, Gonza´lez A, Gonza´lez JM, Gonza´-

lez O, Gon˜i F, Guerrero E, Herna´ndez C,

Herna´ndez A, Herna´ndez JA, Herna´ndez

C, Herna´ndez A, Herna´ndez C, Iglesias P,

Ikobalceta P, Illan F, In˜igo P, Iriarte A,

Irigoyen L, Jaunsolo MA, Lain T, Levy I,

Llorente I, Lo´pez AJ, Lo´pez de la Torre M,

Lo´pez-Guzman A, Losada F, Luna R, Ma-

canas G, Macia C, Maldonado JA, Mangas

MA, Manzanares JM, Maravall FJ, Marco

A, Marco AL, Marin M, Martı´n T, Martı´n

A, Martı´nez MP, Martı´nez FJ, Martı´nez

MA, Matia P, Mato JA, Megia A, Mene´n-

dez E, Meoro A, Merino JF, Mesa J,

Mezquita P, Micalo T, Miralles JM, Molina

MJ, Monereo S, Morales MJ, Moreno A,

Moreno C, Mories T, Mun˜oz L, Mun˜oz M,

Mun˜oz A, Muros T, Nubiola A, Oleaga A,

Oliva´n B, Palacios N, Palomares R, Pazos

F, Pena E, Pen˜aﬁel FJ, Pereyra F, Pe´rez A,

Pesquera C, Picon MJ, Piedrola G, Pinedo

R, Pinie´s J, Pizarro E, Prieto J, Puig M,

Puigdevall V, Pumar A, Recas I, Recio JM,

Reverter J, Ricart W, Rius F, Rivas M,

Rosell R, Rubio JA, Ruiz JM, Sagarra E,

Sales J, Salinas I, Sa´nchez C, Sa´nchez JC,

Santiago P, Santiago L, Sastre J, Segura P,

Simo R, Simo´n I, Solache I, Soriano J, Soto

A, Suarez-Lledo E, Surruca L, Tebar FJ,

Tenes T, Todolı´ J, Tofe S, Tolosa M, Tor-

ras R, Torres A, Vadillo V, Velasco JL,

Vences LF, Vendrell JJ, Vicente A, Vidal JI,

Villa M, Villar A, Zurro J.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Everhart JE, Pettitt DJ, Bennett PH,

Knowler WC: Duration of obesity in-

creases the incidence of NIDDM. Diabetes

41:235–240, 1992

2. American Diabetes Association: Diabetes

mellitus and exercise (Position State-

ment). Diabetes Care 25 (Suppl. 1):S64–

S68, 2002

3. Harding AH, Williams DE, Hennings SH,

Mitchell J, Wareham NJ: Is the association

between dietary fat intake and insulin re-

sistance modiﬁed by physical activity?

Metabolism 50:1186–1192, 2001

4. Boule NG, Haddad E, Kenny GP, Wells

GA, Sigal RJ: Effects of exercise on glycemic

control and body mass in type 2 diabetes

mellitus: a meta-analysis of controlled

clinical trials. JAMA 286: 1218–1227,

2001

Education and

Mortality in Type 2

Diabetes

dults with diabetes tend to have

higher age-adjusted mortality rates

(1,2). Evidence from Britain (3)

suggests that mortality in diabetic adults

is inversely related to socioeconomic sta-

tus. The importance of self-care and

blood glucose control in adults with dia-

betes suggests that education may be an

important factor in long-term health out-

comes, a hypothesis that is also consistent

with evidence in the health economics lit-

erature that education increases the efﬁ-

ciency of investment in health (4,5). The

speciﬁc hypothesis tested here is that

mortality in adults with diabetes is in-

versely related to education.

The data are taken from the National

Health Interview Survey (NHIS) of 1989,

including its diabetes supplement, with

mortality status taken from the NHIS

Multiple Cause of Death Public Use Data

File, including deaths from 1989 to 1995.

The statistical analysis is applied to those

adults who self-reported that they had

been diagnosed with diabetes, excluding

those with apparent type 1 diabetes (age

of onset ⬍30 years, with insulin depen-

dence, and a BMI ⬍27.2 kg/m

for men

and ⬍26.9 for women). The resulting

sample includes 2,387 adults with type 2

diabetes having an average age of 61.8

years.

Logistic regression analysis is used to

explore the relation between mortality

and educational attainment, family in-

come (greater than or less than $20,000

per year in 1989), age, sex, marital status,

race, and duration of diabetes. The statis-

tical analysis is carried out with the svy-

logit procedure in STATA software,

version 7 (STATA, College Station, TX),

using the NHIS sampling weights.

The primary ﬁnding is that mortality

is inversely related to education for dia-

betic adults (OR [95% CI] 0.61 [0.40–

0.93] for college graduation or higher and

0.78 [0.61–0.99] for high school gradu-

ation or higher). Mortality is not signiﬁ-

cantly related to family income, marital

status, or race, but age, male sex, and du-

ration of diabetes have the expected pos-

itive and statistically signiﬁcant associations

with mortality. The ﬁnding that educa-

tion reduces the mortality risk of diabetic

adults is consistent with the predictions

from health investment theory that edu-

cation increases the efﬁciency of health

investment. Education may also be a fac-

tor in the relatively poor health status and

outcomes of adults with diabetes (6).

Acknowledgments—The author thanks

Philip Jacobs for comments, Glenn Harrison

for generous help with the data, and Paula

Veiga for expert research assistance. Any re-

maining errors are mine.

RONALD P. WILDER,

PHD

From the Department of Economics, Moore School

of Business, University of South Carolina, Colum-

bia, South Carolina.

Address correspondence to Ronald P. Wilder,

PhD, Department of Economics, Moore School of

Business, University of South Carolina, Columbia,

SC 29208. E-mail: ronwildr@darla.badm.sc.edu.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Geiss LS, Herman WH, Smith PJ: Mortal-

ity in non-insulin dependent diabetes. In

Diabetes in America. 2nd ed. National

Diabetes Data Group, National Institutes

of Health, National Institute of Diabetes

and Digestive and Kidney Diseases, 1995

(NIH Publ. no. 95-1468)

2. Gu K, Cowie CC, Harris MI: Mortality in

adults with and without diabetes in a na-

tional cohort of the U.S. population,

1971–1993. Diabetes Care 21:1138–

1145, 1998

3. Roper NA, Bilous RW, Kelly WF, Unwin

NC, Connolly VM: Excess mortality in a

population with diabetes and the impact

of material deprivation: longitudinal,

population based study. BMJ 322:1389–

1393, 2001

4. Grossman M, Kaestner R: Effects of edu-

cation on health. In The Social Beneﬁts of

Education. Behrman J, Stacey N, Eds. Ann

Arbor, MI, University of Michigan Press,

1997

5. Kahn ME: Education’s role in explaining

diabetic health investment differentials.

Econ Educ Rev 17:257–266, 1998

6. Harris MI: Health care and health status

and outcomes for patients with type 2 di-

abetes. Diabetes Care 23:854–758, 2000

Stress

Hyperglycemia,

Inﬂammation, and

Cardiovascular Events

ne-third of all individuals with hy-

perglycemia admitted to an urban

general hospital do not have a pre-

vious diagnosis of diabetes; in these pa-

Letters

1650 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

tients, hyperglycemia is a risk factor for

adverse outcomes during acute illness (1).

For example, in patients presenting with

acute myocardial infarction, glucose val-

ues in excess of 6.1–8.0 mmol/l were as-

sociated with a threefold increase in

mortality and a higher risk of heart failure

(2). Control of hyperglycemia with inten-

sive insulin therapy during acute illness

results in marked improvements of clini-

cal outcomes (3).

Dandona et al. (4) put forward a

working hypothesis linking in a feedback

loop of glucose (and possibly other nutri-

ents), insulin, and inﬂammation. Accord-

ing to this paradigm, hyperglycemia has a

proinﬂammatory action that is normally

restrained by the anti-inﬂammatory effect

of insulin secreted in response to that

stimulus. A likely corollary of this para-

digm is that the proinﬂammatory effects

of acute (stress) hyperglycemia may be

implicated in the poor prognosis of hos-

pitalized patients with or without diabe-

tes (particularly death, disability after

acute cardiovascular events, and infec-

tions). We have demonstrated that circu-

lating levels of some proinﬂammatory

cytokines are regulated by glucose levels

in humans (5). Acute hyperglycemia

(⬃15 mmol/l) induced by glucose clamp-

ing in normal subjects and lasting 5 h

determines a signiﬁcant increase of inter-

leukin (IL)-6, IL-18, and tumor necrosis

factor-␣(TNF-␣) circulating levels. These

effects were more sustained in patients

with impaired glucose tolerance, as well

as following consecutive pulses of intra-

venous glucose, and were annulled by

glutathione, implicating an oxidative

mechanism. Interestingly enough, these

cytokines have been implicated in insulin

resistance (TNF-␣and IL-6), atheroscle-

rotic plaque destabilization (IL-18), and

future cardiovascular events (IL-6, IL-18,

and TNF-␣). So, an increased oxidative

stress may be a likely mechanism linking

stress hyperglycemia to cardiovascular

events via an increased cytokine production.

Free fatty acids (FFAs) also induce

proinﬂammatory changes. During illness,

stress increases the concentration of

counterregulatory hormones (mainly glu-

cagon, an epinephrine). Given this back-

ground, it is plausible that in the presence

of high concentrations of both glucose

and FFA, inﬂammation is more promi-

nent. We tested this hypothesis by giving

type 2 diabetic patients two different

meals: a high-carbohydrate meal (high

glucose levels) and a high-fat meal (high

glucose plus high FFA levels). In the latter

condition, the serum concentrations of

cytokines (TNF-␣and IL-6) and adhesion

molecules (intracellular adhesion mole-

cule-1 and vascular cell adhesion mole-

cule-1) were at the highest level (6).

In conclusion, high circulating con-

centrations of glucose and FFAs may ex-

plain, at least in part, the oxidative and

inﬂammatory derangements during acute

illness; insulin may exert its anti-inﬂam-

matory action by ameliorating glucose

and lipid parameters. However, any addi-

tional anti-inﬂammatory effect insulin

may have of its own is welcome. The im-

portance is to recognize that treatment of

even trivial hyperglycemia during acute

illness is fundamental to improve sur-

vival. Because insulin is the best choice to

normalize glucose control during stress,

accepting the premise will inevitably

bring increased insulin use in intensive

care units.

KATHERINE ESPOSITO,

RAFFAELE MARFELLA,

MD, PHD

DARIO GIUGLIANO,

MD, PHD

From the Department of Geriatrics and Metabolic

Diseases, Cardiovascular Research Center, Second

University of Naples, Naples, Italy.

Address correspondence to Dario Giugliano, MD,

PhD, Department of Geriatrics and Metabolic Dis-

eases, Policlinico della Seconda Universita`di

Napoli, Piazza L. Miraglia, 80138 Naples, Italy.

Email: dario.giugliano@unina2.it.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Umpierrez GE, Isaacs SD, Bazargan N,

You X, Thaler LM, Kitabchi AE: Hy-

perglycemia: an independent marker of

in-hospital mortality in patients with

undiagnosed diabetes. J Clin Endocrinol

Metab 87:978–982, 2002

2. Capes SE, Hunt D, Malmberg K, Pathak P,

Gerstein HC: Stress hyperglycemia and

increased risk of death after myocardial

infarction in patients with and without di-

abetes: a systematic overview. Lancet 355:

773–778, 2000

3. Van den berghe G, Wouters P, Weekers F,

Verwaest C, Bruyninckx F, Schetz M,

Vlasselaers D, Ferdinande P, Lauwers P,

Bouillon R: Intensive insulin therapy in

critically ill patients. N Engl J Med 345:

1359–1367, 2001

4. Dandona P, Aljada A, Bandyopadhyay A:

The potential therapeutic role of insulin in

acute myocardial infarction in patients

admitted to intensive care and in those

with unspeciﬁed hyperglycemia. Diabetes

Care 26:516–519, 2003

5. Esposito K, Nappo F, Marfella R, Giugli-

ano G, Giugliano F, Ciotola M, Quagliaro

L, Ceriello A, Giugliano D: Inﬂammatory

cytokine concentrations are acutely in-

creased by hyperglycemia in humans: role

of oxidative stress. Circulation 106:2067–

2072, 2002

6. Nappo F, Esposito K, CiofﬁM, Giugliano

G, Molinari AM, Paolisso G, Marfella R,

Giugliano D: Postprandial endothelial ac-

tivation in healthy subjects and in type 2

diabetic patients: role of fat and carbohy-

drate meals. J Am Coll Cardiol 39:1145–

1150, 2002

Laughter Lowered

the Increase in

Postprandial Blood

Glucose

egative emotions such as anxiety,

fear, and sorrow are known to be

factors that elevate the blood glu-

cose level (1). Conversely, positive emo-

tions such as laughter have been reported

to modify the levels of neuroendocrine

factors involved in negative emotions

(2,3) and to modulate immune function

(3,4). However, there have been no stud-

ies on the effects of laughter on blood glu-

cose level. The purpose of this study was

to clarify changes in the blood glucose

level after laughing episodes in patients

with diabetes.

A 2-day experiment was performed in

19 patients with type 2 diabetes not re-

ceiving insulin therapy (16 men and 3

women, age 63.4 ⫾1.3 years, BMI

23.5 ⫾0.7 kg/m

, HbA

, 7.2 ⫾0.1%

[means ⫾SE]) and 5 healthy subjects

(2 men and 3 women, age 53.6 ⫾3.5

years, BMI 24.3 ⫾1.6 kg/m

, HbA

4.8 ⫾0.1% [means ⫾SE]). On both ex-

perimental days, they consumed the same

500-kcal meal (79.9 g carbohydrate,

21.0 g protein, 7.8 g fat, and 1.0 g ﬁber).

On the ﬁrst day, they attended a monot-

onous lecture (40 min) without humor-

ous content. On the second day, as part

of an audience of 1,000 people attend-

ing MANZAI (a Japanese cross-talk com-

edy) (40 min) in a civic hall, the subjects

laughed. Blood glucose was measured

from the ﬁngertip by enzyme colorimet-

ric assay using a blood glucose self-

measurement apparatus. The subjects

estimated their laughter level on a scale of

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1651

0–5, and most of them considered that

they laughed well (level 4 or 5). Self-

monitoring of blood glucose was per-

formed before food intake (fasting blood

glucose [FBG]) and 2 h after the meal was

started (2-h postprandial blood glucose

[PPBG]).

The results are presented as means ⫾

SE. In the patients, the mean 2-h PPBG

was 6.8 ⫾0.7 mmol/l higher than the

FBG after the lecture and 4.3 ⫾0.8

mmol/l higher after the comedy show.

The difference in the mean increase be-

tween the lecture and comedy show was

2.5 ⫾0.7 mmol/l (P⬍0.005). In the

healthy subjects, the mean increases were

2.0 ⫾0.7 and 1.2 ⫾0.4 mmol/l after the

lecture and comedy show, respectively,

and the difference was 0.8 ⫾0.5 mmol/l

(P⫽0.138).

These results suggest a signiﬁcant

suppression of the increase in 2-h PPBG

by comedy show in patients with diabe-

tes, suggesting that laughter ameliorates

the postprandial glucose excursion in the

presence of insufﬁcient insulin action.

This favorable effect of laughter may in-

clude the acceleration of glucose utiliza-

tion by the muscle motion during the

comedy show. However, it is possible that

positive emotions such as laughter acted

on the neuroendocrine system and sup-

pressed the elevation of blood glucose

level.

In conclusion, the present study elu-

cidates the inhibitory effect of laughter on

the increase in PPBG and suggests the im-

portance of daily opportunities for laugh-

ter in patients with diabetes.

KEIKO HAYASHI,

RN, PHD

TAKASHI HAYASHI,

BAG

SHIZUKO IWANAGA,

KOICHI KAWAI,

MD, PHD

HITOSHI ISHII,

MD, PHD

SHIN’ICHI SHOJI,

MD, PHD

KAZUO MURAKAMI,

PHD

From the

Institute of Community Medicine, Uni-

versity of Tsukuba, Tsukuba, Ibaraki, Japan; the

Foundation for Advancement of International Sci-

ence, Bio-Laboratory, Tsukuba, Ibaraki, Japan; the

Tsukuba Diabetes Center, Kawai Clinic, Tsukuba,

Ibaraki, Japan; the

Department of Endocrinology,

Tenri Hospital, Tenri, Nara, Japan; and the

Institute

of Clinical Medicine, University of Tsukuba,

Tsukuba, Ibaraki, Japan.

Address correspondence to Keiko Hayashi, RN,

PhD, 1-1-1, Tennoudai, Tsukuba, Ibaraki, 305-

8577, Japan. E-mail: khys@sakura.cc.tsukuba.ac.

jp.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Surwit RS, Schneider MS: Role of stress in

the etiology and treatment of diabetes

mellitus. Psychosom Med 55:380–393,

1993

2. Berk LS, Tan SA, Fry WF, Napier BJ,

Lee JW, Hubbard RW, Lewis JE, Eby

WC:Neuroendocrine and stress hormone

changes during mirthful laughter. Am J

Med Sci 298:390–396, 1989

3. Berk LS, Tan SA, Napier BJ, Eby WC: Eu-

stress of mirthful laughter modiﬁes natu-

ral killer cell activity (Abstract). Clin Res

37:115, 1989

4. Takahashi K, Iwase M, Yamashita K, Tat-

sumoto Y, Ue H, Kuratsune H, Shimizu A,

Takeda M: The elevation of natural killer

cell activity induced by laughter in a cross-

over designed study. Int J Mol Med 8:645–

650, 2001

Differential Levels

of ␥-Glutamyl

Transferase Activity

and Apolipoprotein

CIII in Men on Either

Statin or Fibrate

Therapy

igh triglycerides and low HDL cho-

lesterol are frequently associated

with insulin resistance. Statins are

potent hypocholesterolemic drugs, while

ﬁbrates are better at reducing triglycer-

ides and enhancing HDL. However, little

is known of their effects on other param-

eters potentially associated with insulin

resistance, such as ␥-glutamyl transferase

(GGT) (1) and apolipoprotein CIII (2).

In this cross-sectional study, we have

analyzed the biochemical proﬁles of 500

men from Southwestern France as a func-

tion of their hypolipidemic therapy.

Comparison between untreated (n⫽

361) and ﬁbrate (n⫽50)- and statin (n⫽

89)-treated subjects revealed no differ-

ence in their levels of total, LDL, and HDL

cholesterol, apo B, apo AI, and apo E, after

adjustment for age, BMI, alcohol intake,

smoking, and physical activity. This sug-

gests that both therapies are equally effec-

tive at normalizing classic lipid variables.

By contrast, apo CIII was higher in the

statin group (34.3 ⫾13.9 mg/l) than in

both the ﬁbrate (26.7 ⫾10.3, P⬍0.01)

and nontreated (28.7 ⫾10.9 mg/l, P⬍

0.001) groups after adjustment for con-

founders, including triglycerides. Similar

differences were found for Lp B:CIII, a

marker measuring apo CIII associated

with apo B. These observations might in-

dicate an additional beneﬁcial effect of ﬁ-

brates, considering that Lp B:CIII has

been described as an independent predic-

tor of coronary events, better than triglyc-

erides (3).

Most interesting, GGT was two times

lower for ﬁbrates (36.7 ⫾24.4 units/l)

than for the two other groups (67.1 ⫾

71.3 units/l for statins, P⬍0.01 and

58.1 ⫾51.7 units/l in nontreated sub-

jects, P⬍0.01), taking into consideration

potential confounders, including alcohol

intake and triglycerides. GGT is often el-

evated in obese subjects and has been as-

sociated with steatosis, which could be

due to increased effects of insulin in the

liver and could contribute to the develop-

ment of systemic insulin resistance and

hyperinsulinemia in obesity. GGT was re-

ported as one of the independent predic-

tors of the metabolic syndrome and type 2

diabetes (1,4). Moreover, GGT was de-

tected in atheromatous plaques and has

been shown to promote LDL oxidation by

hydrolyzing glutathione into more potent

iron reductants (5). Thus, ﬁbrates may

prove effective in the treatment of dyslip-

idemic diabetic or obese patients, since

they should improve the hepatic suffering

induced by the triglyceride load and also

limit the potential proatherogenic effects

of GGT.

Acknowledgments—This study was sup-

ported by Laboratoires Fournier.

LAURENCE MABILE,

PHD

1,2

JEAN-BERNARD RUIDAVETS,

JOSETTE FAUVEL,

PHD

BERTRAND PERRET,

JEAN FERRIE` RES,

MD, FESC

From the

De´partement d’Epide´miologie, Inserm

U558, Faculte´de Me´decine, Toulouse, France;

In-

serm U563, CHU Toulouse, France; and the

Labo-

ratoire de Biochimie III, CHU Toulouse, France.

Address correspondence to Prof. Jean Ferrie`res,

Inserm U558, Faculte´de Me´decine, 37 alle´es Jules

Guesde, 31 073, Toulouse Cedex 03, France. E-

mail: ferriere@cict.fr.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Rantala AO, Lilja M, Kauma H, Savolainen

MJ, Reunanen A, Kesaniemi YA: Gamma-

glutamyl transpeptidase and the meta-

Letters

1652 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

bolic syndrome. J Intern Med 248:230–

238, 2000

2. Dallongeville J, Meirhaeghe A, Cottel D,

Fruchart JC, Amouyel P, Helbecque N:

Polymorphisms in the insulin response el-

ement of APOC-III gene promoter inﬂu-

ence the correlation between insulin and

triglyceride-rich lipoproteins in humans.

Int J Obes Relat Metab Disord 25:1012–

1017, 2001

3. Sacks FM, Alaupovic P, Moye LA, Cole

TG, Sussex B, Stampfer MJ, Pfeffer MA,

Braunwald E: VLDL, apolipoproteins B,

CIII, and E, and risk of recurrent coro-

nary events in the Cholesterol and Recur-

rent Events (CARE) trial. Circulation 102:

1886–1892, 2000

4. Perry IJ, Wannamethee SG, Shaper AG:

Prospective study of serum gamma-glu-

tamyltransferase and risk of NIDDM. Di-

abetes Care 21:732–737, 1998

5. Paolicchi A, Minotti G, Tonarelli P, Tongi-

ani R, De Cesare D, Mezzetti A, Dominici

S, Comporti M, Pompella A: Gamma-glu-

tamyl transpeptidase-dependent iron re-

duction and LDL oxidation: a potential

mechanism in atherosclerosis. J Investig

Med 47:151–160, 1999

COMMENTS AND

RESPONSES

IGF-1 and

Macrovascular

Complications of

Diabetes

Alternative interpretations of

recently published data

he hypothesis of a detrimental role

for IGF-1 and other growth factors

in the development of vascular dis-

ease (1) has not been consistently sup-

ported by recent studies (2,3). New

evidence suggests that cellular senescence

(4) and impaired vascular endothelial

proliferation, adhesion, and incorpora-

tion play a pivotal role in the development

of macrovascular disease (5). Indeed, re-

cent data on large numbers suggest that

higher IGF-1 bioavailability may protect

against the onset of ischemic heart disease

(6,7) and glucose intolerance (8) and, in

type 2 diabetic patients, may offer im-

proved metabolic control and prevent

vascular complications (9,10). Other po-

tential beneﬁcial actions of IGF-1 in car-

diovascular physiology include increased

nitric oxide synthesis and K

⫹

channel

opening (11,12), and this may explain the

impaired small-vessel function associated

with low IGF-1 levels in patients with car-

diovascular syndrome X (12). By counter-

acting oxidized LDL-induced cytotoxicity

and vascular smooth muscle cell apopto-

sis, IGF-1 may protect against plaque in-

stability and rupture (2). In patients with

acute myocardial infarction, markedly re-

duced IGF-1 values are associated with a

worse outcome (13,14), and recent data

suggest that intramyocardial (15,16) or

vascular (17) gene delivery of growth fac-

tors can improve symptoms and exercise

capacity in patients with coronary or pe-

ripheral vascular disease.

We were therefore surprised by two

recent studies proposing IGF-1 as a me-

diator of harmful vascular effects (18,19).

The authors found an inverse association

between IGF binding protein (IGFBP)-1

and carotid intimal-medial thickness (r⫽

⫺0.135, P⫽0.041) (18), macrovascular

disease (19), and hypertension (19) in pa-

tients with type 2 diabetes. IGFBP-1 is a

minor component of the system of

IGFBPs 1–6, ﬁrst discovered in the

1990s, and regulated by a complex group

of proteases and phosphatases (20). As

IGFBP-1 is generally assumed to inhibit

IGF-1 bioavailability (21), the authors

conclude that IGF-1 exerted harmful vas-

cular effect (18,19).

We propose two different interpreta-

tions of the above data, not necessarily

exclusive. First, the reduced IGFBP-1

concentration observed in situations of

vascular disease (18,19) could result from

activation of a compensatory mechanism

leading to higher IGF-1 bioavailability.

This interpretation is conﬁrmed by the

ﬁnding in hypertensive patients of higher

IGF-1 levels than in control subjects (22),

and of a positive association among

IGF-1, insulin sensitivity, and preserved

vasodilator capacity (23,24). Indeed, in

these patients, IGF-1 was the main inde-

pendent predictor of both coronary re-

serve and insulin sensitivity (23). Second,

alternatively or additionally, one should

consider that the interaction between

IGF-1 and its binding proteins is com-

plex. An adequate serum amount of total

IGFBP-1 is essential for IGF-1’s biological

activity (8,25), independently of its phos-

phorilation status. Moreover, overexpres-

sion of IGFBP-1 attenuates IGF-1’s

growth-promoting actions in vitro but en-

hances them in vivo (26). Thus, a thresh-

old concentration of IGFBP-1 might

enhance, rather than impair, IGF-1’s bio-

activity through a partial agonism-like

action.

Because IGFBPs have been character-

ized only recently and their physiology in

humans is not completely understood, we

think further studies are needed to better

understand the true value of the IGF-1/

IGFBP axis in promoting or protecting

against vascular complications.

ELENA CONTI,

DARIO PITOCCO,

ETTORE CAPOLUONGO,

BSC, PHD

CECILIA ZUPPI,

GIOVANNI GHIRLANDA,

FILIPPO CREA,

MD, FESC, FACC

FELICITA ANDREOTTI,

MD, PHD, FESC

From the

Department of Cardiology, Catholic Uni-

versity, Rome, Italy; the

Department of Diabetolo-

gia, Catholic University, Rome, Italy; and the

Department of Hormonal Chemistry, Rome, Italy.

Address correspondence to Elena Conti, MD, In-

stitute of Cardiology, Catholic University, Via Todi

60, 00181 Rome. E-mail: e_conti02@hotmail.com.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Grant MB, Wargovich TJ, Ellis EA, Cabal-

lero S, Mansour M, Pepine CJ: Localiza-

tion of insulin-like growth factor 1 and

inhibition of coronary smooth muscle cell

growth by somatostatin analogues in hu-

man coronary smooth muscle cells: a po-

tential treatment for restenosis? Circulation

89:1511–1517, 1994

2. Frystyk J, Ledet T, Moller N, Flyvbjerg A,

Orskov H: Cardiovascular disease and in-

sulin-like growth factor 1. Circulation

106:893–895, 2002

3. Janssen JA, Lamberts SWJ: The role of

IGF-1 in the development of cardiovascu-

lar disease in type 2 diabetes mellitus: is

prevention possible? Eur J Endocrinol 146:

467–477, 2002

4. Erusalimsky JD, Fenton M: Further in

vivo evidence that cellular senescence is

implicated in vascular pathophysiology

(Letters). Circulation 106:e144, 2002

5. Tepper OM, Galiano RD, Capla JM, Kalka

C, Gagne PJ, Jacobowitz GR, Levine JP,

Gurtner GC: Human endothelial progen-

itor cells from type II diabetics exhibit

impaired proliferation, adhesion, and in-

corporation into vascular structures. Cir-

culation 106:2781–2786, 2002

6. Juul A, Scheike T, Davidsen M, Gyllen-

borg J, Jorgensen T: Low serum insulin-

like growth factor 1 is associated with

increased risk of ischemic heart disease: a

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1653

population-based case control study. Cir-

culation 106:939–944, 2002

7. Conti E, Crea F, Andreotti F: IGF-1 and

risk of ischemic heart disease. Circulation.

In press

8. Sandhu MS, Heald AH, Gibson JM,

Cruickshank JK, Dunger DB, Wareham

NJ: Circulating concentrations of insulin-

like growth factor-1 and development of

glucose intolerance: a prospective obser-

vational study. Lancet 359:1740–1745,

2002

9. Garay-Sevilla ME, Nava LE, Malacara JM,

Wrobel K, Perez U: Advance glycosylation

end products (AGEs), insulin-like growth

factor-1 (IGF-1) and IGF binding pro-

tein-3 (IGFBP-3) in patients with type 2

diabetes mellitus. Diabetes Metab Res Rev

16:106–113, 2000

10. Janssen JAMIL, Lamberts SWJ: Circulat-

ing IGF-1 and its protective role in the

pathogenesis of diabetic angiopathy. Clin-

ical Endocrinology 52:1–9, 2000

11. Muniyappa R, Walsh MF, Rangi JS, Zayas

RM, Standley PR, Ram JL: Insulin-like

growth factor 1 increases vascular smooth

muscle nitric oxide production. Life Sci-

ence 61:925–931, 1997

12. Conti E, Andreotti F, Sestito A, Riccardi P,

Menini E, Crea F, Maseri A, Lanza GA:

Markedly reduced insulin-like growth

factor-1 associated with insulin resistance

in syndrome X patients. Am J Cardiol 89:

973–975, 2002

13. Conti E, Andreotti F, Sciahbasi A, Ric-

cardi P, Marra G, Menini E, Ghirlanda G,

Maseri A: Markedly reduced insulin-like

growth factor-1 in the acute phase of

myocardial infarction. J Am Coll Cardiol

38:26–32, 2001

14. Friberg L, Werner S, Eggertsen G, Ahnve

S: Growth hormone and insulin-like

growth factor-1 in acute myocardial in-

farction. Eur Heart J 21:1547–1554, 2000

15. Stewart DJ, on behalf of the REVASC

Study Investigators: A phase 2, random-

ized, multicenter, 26-week study to assess

the efﬁcacy and safety of BIOBYPASS

(AdGVVEGF121,10) delivered through

minimally invasive surgery versus maxi-

mum medical treatment in patients with

severe angina, advanced coronary artery

disease, and no options for revasculariza-

tion (Abstract). Circulation 106:2986, 2002

16. Losordo DW, Kawamoto A: Biological re-

vascularization and the interventional

molecular cardiologist: bypass for the

next generation. Circulation 106:3002–

3005, 2002

17. Vale PR, Isner JM, Rosenﬁeld K: Thera-

peutic angiogenesis in critical limb and

myocardial ischemia. J Interv Cardiol 14:

511–528, 2001

18. Leinonen ES, Salonen JT, Salonen RM,

Koistinen RA, Leinonen PJ, Sarna SS,

Taskinen MR: Reduced IGFBP-1 is asso-

ciated with thickening of the carotid wall

in type 2 diabetes. Diabetes Care 25:1807–

1812, 2002

19. Heald AH, Siddals KW, Fraser W, Taylor

W, Kaushal K, Morris J, Young RJ, White

A, Gibson JM: Low circulating levels of

insulin-like growth factor binding pro-

tein-1 (IGFBP-1) are closely associated

with the presence of macrovascular dis-

ease and hypertension in type 2 diabetes.

Diabetes 51:2629–2636, 2002

20. Baxter RC: Insulin-like growth factor

(IGF)-binding proteins: interactions with

IGFs and intrinsic bioactivities. Am J Physiol

Endocrinol Metab 278:E967–E976, 2000

21. Schneider MR, Lahm H, Wu M, Hoeﬂich

A, Wolf E: Transgenic mouse models for

studying the functions of insulin-like

growth factors. FASEB J 14:629–640, 2000

22. Galderisi M, Vitale G, Lupoli G, Barbieri

M, Varricchio G, Carella C, de Divitiis O,

Paolisso G: Inverse association between

free insulin-like growth factor-1 and iso-

volumic relaxation in arterial systemic hy-

pertension. Hypertension 38:240–245, 2001

23. Galderisi M, Caso P, Cicala S, De Simone

L, Barbieri M, Vitale G, de Divitiis O, Pa-

olisso G: Positive association between cir-

culating free insulin-like growth factor-1

levels and coronary ﬂow reserve in arterial

systemic hypertension. Am J Hypertens

15:766–772, 2002

24. Laviades C, Gil MJ, Monreal I, Gonzalez

A, Diez J: Tissue availability of insulin-like

growth factor I is inversely related to in-

sulin resistance in essential hypertension:

effects of angiotensin converting enzyme

inhibition. J Hypertens 16:863–870, 1998

25. Shinobe M, Sanaka T, Nihei H, Nobuhiro

S: IGF-1/IGFBP-1 as an index for discrim-

ination between responder and nonre-

sponder to recombinant human growth

hormone in malnourished uremic pa-

tients on hemodialysis. Nephron 77:29–

36, 1997

26. Modric T, Rajkumar K, Murphy LJ: Thy-

roid gland function and growth in IGF

binding protein-1 transgenic mice. Eur J

Endocrinol 141:149–159, 1999

IGF Binding

Protein-1 and

Carotid Intima-

Media Thickness in

Type 2 Diabetes

Response to Conti et al.

e thank Conti et al. (1) for their

interest in our recent article (2)

and the editor for the opportu-

nity to further discuss the role of the IGF

system in atherosclerosis.

We found an inverse relation between

IGF binding protein-1 (IGFBP-1) and ca-

rotid intima-media thickness (IMT) in

type 2 diabetic patients. In the linear re-

gression model for diabetic subjects with

cardiovascular disease (CVD), IGFBP-1

was one of the main determinants of IMT

(standardized coefﬁcient ␤⫽⫺0.353,

P⫽0.027) together with age, Apo B, and

pulse pressure; other determinants were

diabetes duration, smoking, BMI, sex,

and lipoprotein [Lp(a)]. This model ex-

plained 57.3% of maximum IMT varia-

tion. IGFBP-1 also remained in the linear

regression model for IMT in subjects

without CVD. Our cross-sectional study

setting does not allow the interpretation

to which extent these associations are

causal. In agreement with previous data

(3), IGFBP-1 concentration was inversely

related to insulin resistance examined by

homeostasis model assessment.

On the contrary, after adjusting for

confounders, we found no association be-

tween total IGF-I and either IMT or pres-

ence of CVD. This indicates the relation

between IGFBP-1 and IMT to be indepen-

dent of IGF-I. Our results underline

IGFBP-1 as a marker of insulin resistance

and the metabolic syndrome, but this

cannot be extrapolated to reﬂect an IGF-I

effect on the vasculature. Similarly, Heald

et al. (4) demonstrated an inverse corre-

lation between IGFBP-1, but not IGF-I,

and clinical cardiovascular disease in type

2 diabetic subjects.

Another ﬁnding was the positive cor-

relation of total IGF-I with LDL choles-

terol and Lp(a) and IGFBP-3 with LDL

cholesterol, Lp(a), total cholesterol, and

Apo B after controlling for age, sex, BMI,

and diabetes duration. Thus, in hyper-

cholesterolemic diabetic patients the IGF

axis seems to be activated.

The relation of the IGF system to CVD

is complex and awaits a conclusive an-

swer after over 1,000 published articles

(5). Therefore, we agree with Conti et al.

that further research is needed to under-

stand the entire picture.

EEVA S. LEINONEN,

PEKKA J. LEINONEN,

MD, PHD

MARJA-RIITTA TASKINEN,

MD, PHD

From the

Department of Internal Medicine, Hel-

sinki University Central Hospital, Helsinki, Finland;

and the

Department of Obstetrics and Gynecology,

Helsinki University Central Hospital, Helsinki, Fin-

land.

Address correspondence to Professor Marja-

Letters

1654 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

Riitta Taskinen, Helsinki University Hospital, Bio-

medicum, Haartmaninkatu 8, P.O. Box 700, 00029

Helsinki, Finland. E-mail: mataskin@helsinki.ﬁ.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Conti E, Pitocco D, Capoluongo E, Zuppi

C, Ghirlanda G, Crea F, Andreotti F:

IGF-1 and macrovascular complications

of diabetes: alternative interpretations of

recently published data (Letter). Diabetes

Care 26:1653–1654, 2003

2. Leinonen ES, Salonen JT, Salonen RM,

Koistinen RA, Leinonen PJ, Sarna SS,

Taskinen MR: Reduced IGFBP-1 is asso-

ciated with thickening of the carotid wall

in type 2 diabetes. Diabetes Care 25:1807–

1812, 2002

3. Mohamed-Ali V, Pinkney JH, Panahloo A,

Cwyfan-Hughes S, Holly JM, Yudkin JS:

Insulin-like growth factor binding pro-

tein-1 in NIDDM: relationship with the

insulin resistance syndrome. Clin Endocri-

nol (Oxf) 50:221–228, 1999

4. Heald AH, Siddals KW, Fraser W, Taylor

W, Kaushal K, Morris J, Young RJ, White

A, Gibson JM: Low circulating levels of

IGF binding protein-1 are closely associ-

ated with the presence of macrovascular

disease and hypertension in type 2 diabe-

tes. Diabetes 51:2629–2636, 2002

5. Frystyk J, Ledet T, Møller N, Flyvbjerg A,

Ørskov H: Cardiovascular disease and in-

sulin-like growth factor I. Circulation 106:

893–895, 2002

Metformin in Type 1

Diabetes

Is this a good or bad idea?

he article by Meyer et al. (1) revives

a debate regarding the appropriate-

ness of metformin use for people

with type 1 diabetes. Given the potential

for coexisting lactic acidosis and diabetic

ketoacidosis, how can one justify its use?

Indeed, there was little reason to expect a

beneﬁt in patients who were studied:

nonobese type 1 diabetic subjects with

HbA

⬍9.0% who were taking ⬃0.7

units 䡠kg insulin

⫺1

䡠day

⫺1

. A modest

average reduction of daily insulin require-

ments, 4.3 units, as compared with an in-

crease of 1.7 units for placebo, does not

seem to be worth the trade-off of in-

creased risk for severe hypoglycemia (19

events in metformin group vs. 8 events in

placebo group). There was no differential

effect in terms of HbA

. Only 7 of 31

patients (23%) treated with metformin re-

sponded in terms of a signiﬁcant (20%)

reduction in insulin requirement. Fur-

thermore, it is likely that the incidence of

hypoglycemia would be much greater if

more aggressive metabolic targets of

HbA

had been applied. Despite the fail-

ure to observe diabetic ketoacidosis, the

limited number and short period of ob-

servation does not permit the conclusion

that metformin is safe in ketosis-prone di-

abetic subjects.

We have seen a number of type 1

diabetic patients who have received met-

formin prescriptions by other practitio-

ners. It appears that these prescriptions

were given because of a failure to identify

latent autoimmune diabetes in adults or

because the physician believed that the

potential for insulin dose reduction and

lipid improvement justiﬁed a putative

small risk for diabetic ketoacidosis and

lactic acidosis. The temptation to pre-

scribe metformin is increased because of

the high prevalence of metabolic syn-

drome among U.S. adults (2). Indeed, the

diagnosis of metabolic syndrome can fre-

quently be made in the type 1 diabetic

population. For example, using BMI as a

marker for metabolic syndrome, we ob-

served an average BMI of 27 kg/m

in 343

consecutive subjects with type 1 diabetes;

this means that our average type 1 dia-

betic patient is overweight. A BMI ⱖ30

kg/m

, sufﬁcient to diagnosis obesity, was

observed in 89 of 343 subjects (26%).

Seven of our type 1 diabetic patients had a

BMI ⱖ41 kg/m

. Of these severely obese

subjects, two were receiving metformin

therapy as well as insulin.

The results of the study by Meyer et

al. suggest that a small subset of type 1

diabetic patients beneﬁt in terms of insu-

lin dose reduction when metformin is

added to insulin. Questions about long-

term safety and efﬁcacy in this patient

population remain unanswered. There-

fore, when is it reasonable and defensible

to prescribe metformin in type 1 diabetes?

We suggest that metformin should be

avoided unless the following criteria are

met: 1) insulin resistence is clearly inter-

fering with satisfactory glucose control

despite lifestyle interventions; 2) the risk

of diabetic ketoacidosis is minimized by

an intensive program of insulin, self-

monitoring of blood glucose, urine ke-

tone measurement if blood glucose

exceeds 300 mg/dl, and regular medical

supervision; 3) the patient receives coun-

seling so that he or she understands the

potential risk for lactic acidosis; and 4)

efﬁcacy is frequently evaluated to justify

continued use of metformin.

J. DAVID FAICHNEY,

MD, FACP

PHILIP W. TATE,

From the Center for Diabetes and Endocrinology,

Saint Mary’s Mercy Medical Center, Grand Rapids,

Michigan.

Address correspondence to J. David Faichney,

Center for Diabetes and Endocrinology, Saint Mary’s

Mercy Medical Center, 300 Lafayette SE, Suite 2045,

Grand Rapids, MI 49503. E-mail: faichned@trinity-

health.org.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Meyer L, Bohme P, Delbachian I, Lehert P,

Cugnardey N, Drouin P, Guerci B: The

beneﬁts of metfomin therapy during con-

tinuous subcutaneous insulin infusion

treatment of type 1 diabetic patients. Di-

abetes Care 25:2153–2158, 2002

2. Ford ES, Giles WH, Dietz WH: Prevalence

of the metabolic syndrome among US

adults: ﬁndings of the third National

Health and Nutrition Examination Sur-

vey. JAMA 287:356–359, 2002

Metformin and

Insulin in Type 1

Diabetes

The ﬁrst step

he insulin-sparing effect of met-

formin and its positive effect on glu-

cose metabolism is now well

documented in insulin-treated type 2 di-

abetic patients, although the mechanisms

of these effects are not overly clear (1). In

insulin-treated type 1 diabetic patients,

the use of metformin has only been as-

sessed in a few studies and, surprisingly,

the same insulin-sparing effect has been

found despite some methodological difﬁ-

culties. However, the clinical interest of

metformin in the treatment of type 1 dia-

betes has remained questionable. For the

ﬁrst time, we have shown in a randomized

double-blind trial that metformin has an

insulin-sparing effect in type 1 diabetic

patients (even if the studied population

was not clinically insulin resistant) (2).

The selected patients were C-peptide

negative after intravenous glucagon injec-

tion and were not identiﬁed as having

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1655

metabolic syndrome. The patients were

moderately overweight (BMI close to 25

kg/m

), and their daily insulin needs were

not greater than that seen for other type 1

diabetic patients. The insulin-sparing ef-

fect of metformin was obtained for basal

insulin needs but not for prandial needs.

Although no deﬁnitive explanation could

be given, it is now clear that the use of

metformin may be appropriate for some

type 1 diabetic patients.

Our study was designed to assess the

insulin-sparing effect of metformin in in-

sulin-treated type 1 diabetic patients, and

this end was achieved. However, it was

not designed to give clinical or biological

criteria of indications associated with

metformin use and insulin. We found that

a subset of patients had a reduction of at

least 20% in insulin requirements with

stable and satisfactory glucose control,

but a backward logistic regression did not

allow prediction criteria of a good re-

sponse to be deﬁned. Thus, an expanded

study using selection criteria identiﬁed in

our previous study coupled with a larger

range of BMI or bodyweight and a larger

range of daily insulin doses in selected

patients must be performed to determine

therapeutic indications of this association

We believe that metformin use could

be beneﬁcial in type 1 diabetic patients

who are overweight or obese, are receiv-

ing large doses of insulin, and have an

HbA

⬎8%. The use of metformin could

be very useful in these patients but must

be carried out under clinical and biologi-

cal supervision to avoid complications

such as lactic acidosis, which remains ex-

ceptional in our experience of type 1 and

in type 2 diabetic patients.

LAURENT MEYER,

BRUNO GUERCI,

MD, PHD

From the Service de Diabe´tologie, Maladies

Me´taboliques & Maladies de la Nutrition, CIC-

INSERM, Hoˆpital Jeanne d’Arc, Centre Hospitalo-

Universitaire de Nancy, Toul Cedex, France.

Address correspondence to Laurent Meyer, Ser-

vice de Diabe´tologie, Maladies Me´taboliques & Mal-

adies de la Nutrition, CIC-INSERM, Hoˆpital Jeanne

d’Arc, Centre Hospitalo-Universitaire de Nancy,

B.P. 303, 54201 Toul Cedex, France. E-mail:

laurentmeyer@hotmail.com.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Wulffele´MG, Kooy A, Lehert P, Bets D,

Ogterop JC, Borger van der Burg B,

Donker AJ, Stehouwer CD: Combination

of insulin and metformin in the treatment

of type 2 diabetes. Diabetes Care 25:2133–

2140, 2002

2. Meyer L, Bo¨hme P, Delbachian I, Lehert P,

Cugnardey N, Drouin P, Guerci B: The

beneﬁts of metformin therapy during

continuous subcutaneous insulin infu-

sion treatment of type 1 diabetic patients.

Diabetes Care 25:2153–2158, 2002

Prospective Relation

of C-Reactive Protein

With Type 2

Diabetes

Response to Han et al.

e read with great interest the ar-

ticle by Han et al., which was re-

cently published in Diabetes Care

(1). The authors studied the association

between baseline levels of C-reactive pro-

tein (CRP) and the 6-year incidence of the

metabolic syndrome and type 2 diabetes

in Mexican subjects. They found an odds

ratio (OR) of 4.1 (95% CI 2.1–8.0) for

developing the metabolic syndrome and

an OR of 5.4 (2.2–13.4) for incident type

2 diabetes among women in the highest

compared with the lowest tertile of CRP.

ORs were adjusted for age, smoking,

physical activity, and alcohol intake. In

men, no such association was found. Ad-

ditional adjustment for BMI slightly low-

ered the ORs; adjustment for waist-to-hip

ratio (WHR) did not.

We have studied baseline CRP levels

and their relation to incident type 2 dia-

betes in an age-, sex-, and glucose-

stratiﬁed sample of the Hoorn Study, a

population-based cohort study of glucose

tolerance among Caucasian people (2).

The study methods and follow-up dura-

tion are closely similar to the Mexico City

Diabetes Study (MCDS) described by Han

et al., except that our subjects were ⬃15

years older at baseline. Glucose tolerance

status was assessed by a 75-g oral glucose

tolerance test at baseline and after 6.4

years of follow-up. CRP was measured in

plasma by high-sensitive enzyme-linked

immunosorbent assay, while information

was available on BMI, WHR, smoking,

and physical activity. Of the 140 men and

139 women who had follow-up measure-

ments and were free of diabetes at base-

line, 17.8% of the men and 20.9% of the

women developed diabetes. In contrast to

the ﬁndings of Han et al., we did not ob-

serve an association between baseline

CRP levels and incident diabetes in

women, while in men we found an OR of

3.0 (95% CI 1.0–9.3) in the highest com-

pared with the lowest tertile of CRP, after

adjustment for age (Table 1). Further ad-

justment for BMI, smoking, or physical

activity did not materially change these

results, while adding WHR to the model

substantially lowered the OR in men.

Thus, in the Hoorn Study, CRP is not a

very strong determinant of the develop-

ment of type 2 diabetes, in contrast to

WHR and impaired glucose metabolism

(3). We realize that our study sample of

279 subjects is much smaller than the

study presented by Han et al. (n⫽1,244).

The relationship in men, however, was

strong, whereas in the MCDS it was

strong only in women. In the previous

prospective study by Barzilay et al. (4) an

OR of 2.0 (1.4–2.9) was found when ex-

treme quartiles of CRP were compared.

They did not analyze men and women or

black and white subjects separately.

Pradhan et al. (5) found an OR of 4.2

(1.5–12.0) in women. Both studies did

not take WHR into account. Ethnicity

Table 1—Relative risk associated with high C-reactive protein (second and third tertile com-

pared with the ﬁrst tertile) for developing type 2 diabetes after 6.4 years of follow-up in the

Hoorn Study

Model

Men Women

2nd tertile 3rd tertile 2nd tertile 3rd tertile

1 Age 1.4 (0.4–4.8) 3.0 (1.0–9.3) 0.7 (0.3–2.1) 1.1 (0.4–3.0)

2 Model 1 ⫹BMI 1.4 (0.4–4.8) 3.0 (1.0–9.3) 0.6 (0.2–1.8) 0.9 (0.3–2.5)

3 Model 1 ⫹WHR 1.4 (0.4–4.8) 1.9 (0.6–6.4) 0.6 (0.2–1.7) 0.9 (0.3–2.5)

4 Model 1 ⫹smoking 1.3 (0.4–4.7) 2.9 (0.9–9.2) 0.7 (0.3–2.1) 1.2 (0.4–3.2)

5 Model 1 ⫹physical activity 1.4 (0.4–4.7) 3.0 (1.0–9.3) 0.7 (0.3–2.1) 1.1 (0.4–3.0)

Data are OR (95% CI).

Letters

1656 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

may also play an important role in ex-

plaining the inconsistent results.

In conclusion, we suggest caution in

the interpretation of the results because of

inconsistent ﬁndings, in particular be-

tween sexes. It is unclear why inﬂamma-

tion would be important in the patho-

genesis of type 2 diabetes only in women

or only in men.

MARIEKE B. SNIJDER,

MSC

JACQUELINE M. DEKKER,

PHD

MARJOLEIN VISSER,

PHD

COEN D.A. STEHOUWER,

MD, PHD

1,2

JOHN S. YUDKIN,

1,3

LEX M. BOUTER,

PHD

ROBERT J. HEINE,

MD, PHD

1,4

GIEL NIJPELS,

MD, PHD

JACOB C. SEIDELL,

PHD

1,5

From the

Institute for Research in Extramural Med-

icine, VU University Medical Center, Amsterdam,

the Netherlands; the

Department of Internal Med-

icine, VU University Medical Center, Amsterdam,

the Netherlands; the

Department of Medicine, Di-

abetes and Cardiovascular Disease Academic Unit,

University College London Medical School, Lon-

don, U.K.; the

Department of Endocrinology, VU

University Medical Center, Amsterdam, the Nether-

lands; and the

Department of Nutrition and Health,

Faculty of Earth and Life Sciences, Vrije Universiteit,

Amsterdam, the Netherlands.

Address correspondence to Marieke B. Snijder,

MSC, Institute for Research in Extramural Medicine,

VU University Medical Center, Van der Boechorst-

straat 7, 1081 BT Amsterdam, The Netherlands. E-

mail: mb.snijder.emgo@med.vu.nl.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Han TS, Sattar N, Williams K, Gonzalez-

Villalpando C, Lean ME, Haffner SM: Pro-

spective study of C-reactive protein in

relation to the development of diabetes

and metabolic syndrome in the Mexico

City Diabetes Study. Diabetes Care 25:

2016–2021, 2002

2. Mooy JM, Grootenhuis PA, de Vries H,

Valkenburg HA, Bouter LM, Kostense PJ,

Heine RJ: Prevalence and determinants of

glucose intolerance in a Dutch Caucasian

population: the Hoorn Study. Diabetes

Care 18:1270–1273, 1995

3. de Vegt F, Dekker JM, Jager A, Hienkens

E, Kostense PJ, Stehouwer CD, Nijpels G,

Bouter LM, Heine RJ: Relation of impaired

fasting and postload glucose with incident

type 2 diabetes in a Dutch population: the

Hoorn Study. JAMA 285:2109–2113, 2001

4. Barzilay JI, Abraham L, Heckbert SR,

Cushman M, Kuller LH, Resnick HE,

Tracy RP: The relation of markers of in-

ﬂammation to the development of glucose

disorders in the elderly: the Cardiovas-

cular Health Study. Diabetes 50:2384–

2389, 2001

5. Pradhan AD, Manson JE, Rifai N, Buring

JE, Ridker PM: C-reactive protein, inter-

leukin 6, and risk of developing type 2 dia-

betes mellitus. JAMA 286:327–334, 2001

Prospective Relation

of C-Reactive Protein

With Type 2 Diabetes

Response to Snijder et al.

e thank Snijder et al. (1) for their

interest in our article (2) and for

sharing their data relating C-re-

active protein (CRP) to the risk of diabetes

development in the Hoorn Study. We

readily agree that it is unclear why inﬂam-

mation might be more important to the

pathogenesis of type 2 diabetes in women

compared with men or indeed vice versa.

We have reanalyzed our CRP data using

the same covariates as Snijder et al. and

essentially our results are unchanged—

the relation between CRP and diabetes re-

mains strong in women, even after

inclusion of waist-to-hip ratio (WHR),

but is absent in men (Table 1). Interest-

ingly, on reanalysis of similarly relevant

data from the Insulin Resistance Athero-

sclerosis Study (IRAS) (3), a study of sim-

ilar size to the Mexico City Diabetes Study

(MCDS), the relation between baseline

CRP and diabetes development was also

stronger in women compared with men

using the same covariate adjustments

(data not shown). Moreover, in both the

MCDS and IRAS, adjustment for BMI at-

tenuated the link between CRP and dia-

betes more so than WHR.

One possibility of the divergent re-

sults could be the younger baseline age

(⬃46 years) in MCDS in comparison to

that of the Hoorn study (⬃61 years). In

this respect it is important to consider the

additional potential confounding inﬂu-

ence of hormonal replacement therapy

(HRT) in women. HRT use increases CRP

concentrations in women (4), but para-

doxically recent data from the Heart and

Estrogen/Progestin Replacement Study

suggest that HRT reduces risk of diabetes

development by 35% (5). In MCDS, only

4% of women were on HRT at baseline

and hormonal use did not confound the

link between CRP and diabetes. However,

because of higher age, a potentially much

higher proportion of women in the Hoorn

Study may have been on HRT at baseline,

which could thus disguise any potential

association between CRP and diabetes

risk.

We acknowledge that our inability to

ﬁnd an association between CRP and di-

abetes risk in men may simply be due to a

lack of power. A recent relevant study (6)

from the West of Scotland Coronary Pre-

vention Study (WOSCOPS) group that

included 127 new cases (predominantly

Caucasian) of diabetes reported a strong

association between baseline CRP and di-

abetes risk independent of BMI, blood

pressure, smoking, and fasting lipids and

glucose concentrations. Alternatively and

as Snijder et al. suggest, ethnicity may also

play a role in explaining inconsistent

results.

Clearly, further studies are needed to

examine the relation between inﬂamma-

Table 1—ORs (with 95% CIs) associated with high C-reactive protein (second and third tertile

compared to ﬁrst tertile) for developing type 2 diabetes after 6.5 years of follow-up in the

Mexico City Diabetes Study

Model

Men

(35 incident cases/n⫽460)

Women

(54 incident cases/n⫽672)

Second tertile

(0.91–1.64 ng/ml)

Third tertile

(⬎1.64 ng/ml)

Second tertile

(1.24–2.16 ng/ml)

Third tertile

(⬎2.16 ng/ml)

1 Age 0.7 (0.3–1.7) 1.0 (0.5–2.3) 3.1 (1.2–7.9)* 5.7 (2.3–13.9)†

2 Model 1 ⫹BMI 0.6 (0.2–1.4) 0.8 (0.4–1.9) 2.2 (0.9–5.8)* 3.1 (1.2–8.1)*

3 Model 1 ⫹WHR 0.6 (0.2–1.5) 0.9 (0.4–2.1) 3.0 (1.2–7.7)* 5.4 (2.2–13.3)†

4 Model 1 ⫹smoking 0.7 (0.3–1.7) 1.0 (0.5–2.3) 3.1 (1.2–8.0)* 5.7 (2.3–14.1)†

5 Model 1 ⫹physical

activity

0.7 (0.3–1.6) 0.9 (0.4–2.0) 2.7 (1.0–7.1)* 5.3 (2.2–13.2)†

The main effect of gender is signiﬁcant (P⬍0.05) for each model. Gender-CRP tertile interaction: *P⬍0.05,

†P⬍0.01.

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1657

tory markers and diabetes risk. It would

be useful if such studies were to report

associations separately in men and

women in order to enable identiﬁcation of

any potential gender differences.

NAVEED SATTAR,

MD, PHD

KEN WILLIAMS,

THANG S. HAN,

MD, PHD

CLICERIO GONZALEZ-VILLALPANDO,

MICHAEL E.J. LEAN,

MD, FRCP

STEVEN M. HAFFNER,

MD, MPH

From the

University Department of Pathological

Biochemistry, Glasgow Royal Inﬁrmary, Glasgow

U.K.; the

Department of Medicine #7873, Univer-

sity of Texas Health Science Center at San Antonio,

San Antonio, Texas;

Addenbrooke’s Hospital,

Cambridge University Medical School, Cambridge,

U.K.; the

Center de Estudios in Diabetes, Mexico

City, Mexico; and the

University Department of

Human Nutrition, Glasgow Royal Inﬁrmary, Glas-

gow, U.K.

Address correspondence to Steven M. Haffner,

MD, Department of Medicine #7873, University of

Texas Health Science Center at San Antonio, San

Antonio, TX 78229-3900. E-mail: haffner@

uthscsa.edu.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Snijder MB, Dekker JM, Visser JM, Ste-

houwer CDA,Yudkin JS, Bouter LM,

Heine RJ, Nijpels G, Seidell JC: Prospec-

tive relation of C-reactive protein with

type 2 diabetes (Letter). Diabetes Care 26:

1656–1657, 2003

2. Han TS, Sattar N, Williams K, Gonzalez-

Villalpando C, Lean ME, Haffner SM: Pro-

spective study of C-reactive protein in

relation to the development of diabetes

and metabolic syndrome in the Mexico

City Diabetes Study. Diabetes Care 25:

2016–2021, 2002

3. Festa A, D’Agostino R Jr, Tracy RP,

Haffner SM: Elevated levels of acute-

phase proteins and plasminogen activator

inhibitor-1 predict the development of

type 2 diabetes: the insulin resistance ath-

erosclerosis study. Diabetes 51:1131–

1137, 2002

4. Manning PJ, Sutherland WH, Allum AR,

de Jong SA, Jones SD: Effect of hormone

replacement therapy on inﬂammation-

sensitive proteins in post-menopausal

women with type 2 diabetes. Diabet Med

19:847–852, 2002

5. Kanaya AM, Herrington D, Vittinghoff E,

Lin F, Grady D, Bittner V, Cauley JA, Bar-

rett-Connor E: Glycemic effects of post-

menopausal hormone therapy: the Heart

and Estrogen/progestin Replacement

Study: a randomized, double-blind, pla-

cebo-controlled trial. Ann Intern Med 138:

1–9, 2003

6. Freeman DJ, Norrie J, Caslake MJ, Gaw A,

Ford I, Lowe GD, O’Reilly DS, Packard CJ,

Sattar N: C-reactive protein is an indepen-

dent predictor of risk for the development

of diabetes in the West of Scotland Coro-

nary Prevention Study. Diabetes 51:

1596–600, 2002

A Multivariate

Logistic Regression

Equation to Screen

for Diabetes

Response to Tabaei and Herman

e read the article by Tabaei and

Herman (1) in the November

2002 issue of Diabetes Care with

great interest and feel that it covers an

exciting and vital area of diabetes re-

search. We applaud the authors’attempts

to improve current standards of diabetes

screening and the journal’s willingness to

publish important new ﬁndings in the

ﬁeld. However, when we reviewed the ar-

ticle, we found the results difﬁcult to re-

produce, especially as there were no

numerical examples provided in the text.

Using the precise formula given in the ar-

ticle, our calculations produced negative

probability values. We would therefore

like to take this opportunity to point out a

source of possible error in the equation

and clarify the logistic regression model.

The logistic regression model and the

“logit”model can be written as:

ln关P/共1⫺P)] ⫽␤

0⫹␤

iXi⫹e (2)

where X ⫽␤

⫹␤

are the coefﬁcients

(parameters) for the linear predictor and

represents the values of the explanatory

variables (we have assumed that these co-

efﬁcients have been correctly estimated in

the article); ln is the natural logarithm,

log

exp

, where exp ⫽2.71828. . . ; Pis the

probability that the event “previously un-

diagnosed diabetes”occurs; P/(1 ⫺P)is

the “odds ratio”; and ln[P/(1 ⫺P)] is the

log odds ratio, or “logit.”

In contrast to the simple linear prob-

ability model (Y ⫽a⫹bX ⫹e), the logit

distribution constrains the estimated

probabilities to lie between 0 and 1. For

instance, the estimated probability is:

P⫽[exp(␤0⫹␤

iXi)]/[1 ⫹exp(␤0⫹␤

iXi)]

P⫽1/[1 ⫹exp (⫺␤0⫺␤

iXi)]

For a 45-year-old man with a BMI of

29 kg/m

, a plasma glucose level of 125

mg/dl, and a postprandial time of 3 h, the

probability calculated with the published

formula would be P⫽⫺0.0555 (i.e. P⬍

0). This result does not match the require-

ment for probabilities to lie between 0 and

1. For the same case, but with the cor-

rected formula, the probability of being

previously undiagnosed for diabetes is

P⫽0.04996.

Details of the calculation with the

formula published in the article:

P⫽1/[1 ⫺exp(⫺X)]

X⫽⫺10.0382 ⫹0.0331*(age ⫽45

years) ⫹0.0308*(plasma glucose ⫽125

mg/dl) ⫹0.25*(postprandial time ⫽3h)

⫹0.562*(sex ⫽0) ⫹0.0346*(BMI ⫽29

kg/m

). Xⴝⴚ2.9453 and P⫽1/[1 ⫺

exp(⫺X)] ⫽1/(1 ⫺exp(2.9453) ⫽1/(1 ⫺

19.01636644) ⫽1/⫺18.01636644. Pⴝ

ⴚ0.0555.

Details of the calculation with the

corrected formula:

P⫽1/[1 ⫹exp(⫺X)]

X⫽⫺2.9453; P⫽1/[1 ⫹exp(⫺X)] ⫽

1/(1 ⫹exp[2.9453]) ⫽1/(1 ⫹

19.01636644) ⫽1/20.01636644; Pⴝ

0.04996.

Therefore, we would suggest that the

following equations should be used:

X⫽⫺10.0382 ⫹0.0331*(age) ⫹0.0308*

(plasma glucose) ⫹0.25* (postprandial

time) ⫹0.562 (if female) ⫹0.0346* (BMI)

and P⫽1/[1 ⫹exp(⫺X)]

We hope these comments serve to

clarify this logistic regression equation to

screen for diabetes and, in conjunction

with the valuable work of our colleagues

Drs. Tabaei and Herman, help expand the

knowledge base in this key area of diabe-

tes research.

STE

´PHANE ROZE,

MSC

ANDREW J. PALMER,

BSC, MBBS

WILLIAM J. VALENTINE,

PHD

Letters

1658 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

From the Center for Outcomes Research (CORE),

Basel, Switzerland.

Address correspondence to Ste´phane Roze,

CORE: Center for Outcomes Research, St. Johanns-

Ring 139, 4056 Basel, Switzerland. E-mail: roze@

thecenter.ch.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Tabaei BP, Herman WH: A multivariate

logistic regression equation to screen for

diabetes. Diabetes Care 25:1999–2003,

2002

2. Hosmer D, Lemeshow S: Applied Logistic

Regression. New York, Wiley, 1989

A Multivariate

Logistic Regression

Equation to Screen

for Diabetes

Response to Roze, Palmer, and

Valentine

e would like to thank Roze,

Palmer, and Valentine (1) for

their careful reading of our arti-

cle (2) entitled “A multivariate logistic

regression to screen for diabetes: develop-

ment and validation.”The estimated

probability formula for the logistic regres-

sion model was indeed misprinted with a

minus sign. The correct estimated proba-

bility formula is:

P⫽1/(1 ⫹e

⫺x

), where x ⫽

⫺10.0382 ⫹[0.0331 (age in years) ⫹

0.0308 (random plasma glucose in mg/

dl) ⫹0.2500 (postprandial time assessed

as0to8⫹hours) ⫹0.5620 (if female) ⫹

0.0346 (BMI)].

BAHMAN P. TABAEI,

MPH

WILLIAM H. HERMAN,

MD, MPH

1,2

From the

Department of Internal Medicine, Uni-

versity of Michigan Health System, Ann Arbor,

Michigan; and the

Department of Epidemiology,

University of Michigan Health System, Ann Arbor,

Michigan.

Address correspondence to William H. Herman,

MD, MPH, University of Michigan Health System,

Division of Endocrinology and Metabolism, 1500

East Medical Center Dr., 3920 Taubman Center,

Ann Arbor, MI 48109-0354. E-mail: wherman@

umich.edu.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. Roze S, Palmer AJ, Valentine WJ: A mul-

tivariate logistic regression equation to

screen for diabetes (Letter). Diabetes Care

26:1658–1659, 2003

2. Tabael BP, Herman WH: A multivariate

logistic regression equation to screen for

diabetes. Diabetes Care 25:1999–2003,

2002

␤2-Adrenergic

Receptor

Polymorphism and

Obesity in Type 2

Diabetes

Response to van Tilburg et al.

e read with interest the observa-

tion letter by van Tilburg et al.

(1) reporting that the Glu27Gln

polymorphism of the ␤2-adrenergic re-

ceptor (B2ADR) has no important effect

on BMI, metabolic control, or plasma lip-

ids. However, the authors did not men-

tion the physical activity and diet regimen

in their study. B2ADR is a major lipolytic

receptor in human adipocytes, activated

by catecholamines especially during a

weight reduction regimen. Meirhaeghe et

al. (2) reported that B2ADR Glu27Gln

polymorphism is signiﬁcantly associated

with obesity in individuals who do not

have regular physical activity, while no

effect of such polymorphism is found in

those who do have physical activity, and

suggested that obese individuals with

B2ADR Glu27Gln genotype may beneﬁt

from physical activity to reduce their body

weight. We also demonstrated the same as-

sociation between another B2ADR poly-

morphism (Arg16Gly) and obesity in

individuals treated with a combined low-

energy diet and exercise regimen (3). There-

fore, the authors should have further

divided the subjects according to the degree

of their physical activity and analyzed the

data, because body weight and body fat or

metabolic control are inﬂuenced by physi-

cal activity and dietary changes, especially

in individuals with such polymorphisms.

KEIJI YOSHIOKA,

TOSHIHIDE YOSHIDA,

TOSHIKAZU YOSHIKAWA,

From the

Department of Diabetes and Endocrinol-

ogy, Matsushita Memorial Hospital, Moriguchi,

Osaka, Japan; and the

Department of Diabetes, En-

docrinology, and Metabolism, Kyoto Prefectural

University of Medicine, Koyto, Japan.

Address correspondence to Keiji Yoshioka. De-

partment of Diabetes and Endocrinology, Matsushi-

ta Memorial Hospital, 5-55, Sotojma-cho, Moriguchi,

Osaka, Japan. E-mail: yoshik@mue.biglobe.ne.jp.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. van Tilburg JH, Wijmenga C, van Bakel H,

Rozeman L, Pearson PL, van Haeften TW:

Relationship of ␤2-adrenergic receptor

polymorphism with obesity in type 2

diabetes (Letter). Diabetes Care 26:251–

252, 2003

2. Meirhaeghe A, Helbecque N, Cottel D,

Amouyel P: ␤

-adrenoceptor gene poly-

morphism, body weight, and physical ac-

tivity. Lancet 353:896, 1999

3. Sakane N, Yoshida T, Umekawa T, Ko-

gure A, Kondo M: ␤

-adrenoceptor gene

polymorphism and obesity. Lancet 353:

1976, 1999

␤2-Adrenergic

Receptor

Polymorphism and

Obesity in Type 2

Diabetes

Response to Yoshioka, Yoshida,

and Yoshikawa

e recently reported that our

group (n⫽502) of type 2 dia-

betic subjects did not reveal sta-

tistically signiﬁcant differences in BMI

with respect to the three groups of carriers

of ␤2-adrenergic receptor polymor-

phisms (1). Meirhaghe et al. (1) report

that in men not participating in physical

activity, Gln27Gln carriers had a higher

BMI than the Glu27 (combined heterozy-

gous and homozygous) carriers (27.2 ⫾

0.4 vs. 25.2 ⫾0.3 kg/m

), while no effect

on BMI was found in the men not partic-

ipating in physical activity.

We have not collected data on the

physical activity of our subjects. How-

ever, taking into account that our subjects

had a mean age of 61 ⫾9 years (roughly

10 years older than the subjects reported

by Meirhaghe et al.), we suspect that the

vast majority of them do not participate in

much physical activity. In any case, their

physical activity is presumably less than

that of the subjects studied by Meirhaghe

et al. Much to our surprise, we actually

Letters

DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003 1659

ﬁnd the opposite, which is a lower (and

not a higher!) BMI (27.2 ⫾3.8) in the

Gln27Gln carriers compared with the

others (27.9 ⫾4.4 for Gln27Glu and

28.2 ⫾4.0 kg/m

for Glu27Glu), al-

though the differences did not attain sta-

tistical signiﬁcance. These ﬁndings were

similar for men and women when taken

together or separately.

Obviously, there are two major differ-

ences between these two studies. The sub-

jects in our study (2) had type 2 diabetes

and were considerably older. Whether

these two differences explain the diver-

gent ﬁndings of the study of Meirhaghe et

al. and our study is uncertain. However,

the suggestion by van Tilburg et al. that, if

anything, the BMI was lower (and not

higher) in the Gln27Gln27 group might

point to a chance ﬁnding in the studies by

Meirhaghe et al. As for the observations

regarding the Arg16Gly polymorphism,

we have not studied this polymorphisms

in our group of subjects.

It has become more and more clear

that association studies of a particular

polymorphism with a certain physiologi-

cal parameter are often hard to replicate;

this necessitates the need to perform rep-

lication studies before a ﬁnding can be

regarded as truly positive.

JONATHAN H. VAN TILBURG,

PHD

CISCA WIJMENGA,

PHD

TIMON W. VAN HAEFTEN,

From the

Department of Biomedical Genetics, Uni-

versity Medical Center, Utrecht, the Netherlands;

and the

Department of Internal Medicine, Univer-

sity Medical Center, Utrecht, the Netherlands.

Address correspondence to Timon W. van

Haeften, MD, Department of Internal Medicine, G

02.228, University Medical Center, P.O. Box 85500,

3508 GA Utrecht, The Netherlands. E-mail: t.w.

vanhaeften@azu.nl.

●●●●●●●●●●●●●●●●●●●●●●●

References

1. van Tilburg JH, Wijmenga C, van Bakel H,

Rozeman L, Pearson PL, van Haeften TW:

Relationship of ␤2-adrenergic receptor

polymorphism with obesity in type 2

diabetes (Letter). Diabetes Care 26:251–

252, 2003

2. Meirhaghe A, Helbecque N, Cottel D,

Amouyel P: B

-adrenoceptor gene poly-

morphism, body weight, and physical ac-

tivity. Lancet 353:896, 1999

Letters

1660 DIABETES CARE,VOLUME 26, NUMBER 5, MAY 2003

Association of C-Reactive Protein with Risk of Developing Type 2 Diabetes Mellitus, and Role of Obesity and Hypertension: A Large Population-Based Korean Cohort Study

Article

Full-text available

Mar 2019

This study was undertaken to assess the associations of C-reactive protein (CRP) with incident type-2 diabetes mellitus (T2DM) and to determine the joint effect of obesity and hypertension on them in the large-scale population-based Korean cohort of the Korean Genome and Epidemiology study (KoGES). We included 22,946 men and women from 11 rural communities at baseline (2005–2011). Epidemiological data and blood samples were collected. Incident physician-diagnosed T2DM cases (130 men and 148 women) were self-reported or based on fasting glucose ≥126 mg/dL or HbA1c level ≥6.5% during a median follow-up of 3.0 years (58,916 person-years) between 2007 and 2014. After multivariate adjustment for T2DM risk factors, the hazard ratios for developing T2DM in the highest CRP tertile (T3), compared with the lowest (T1), was 2.80 (1.73–4.52; p for trend <0.0001) in women and 1.67 (1.00–2.45; p for trend 0.02) in men. The associations between CRP and incident T2DM were more prominent among the older group (≥50 years). And CRP and its combination with obesity and hypertension were associated with increased risk of T2DM. In conclusion, we found positive associations between CRP and incident T2DM in a large population-based Korean cohort.

The Interaction of Helicobacter pylori Infection and Type 2 Diabetes Mellitus

Article

Full-text available

Feb 2019

Helicobacter pylori is one of the most common human pathogens that can cause gastrointestinal (GI) disorders, including simple gastritis, gastric ulcer, and malignant gastritis. In some cases, such as immunodeficiency and underlying diseases, it can be problematic as opportunistic infections. Diabetes mellitus (type 2) (T2DM) is one of the H. pylori underlying diseases. Since GI problems are observed in diabetic patients, it is necessary to treat H. pylori infection. In this review, we aimed to evaluate the possible relationship between H. pylori and T2DM according to epidemiological surveys of 70 studies retrieved from databases, including Scopus, PubMed, and Google Scholar about the relationship between H. pylori and T2DM, and discuss the reported background mechanisms of this correlation. According to the results of our study, the different studies have shown that H. pylori is more prevalent in Type 2 diabetic patients than healthy individuals or nondiabetic patients. The reason is development of H. pylori infection-induced inflammation and production of inflammatory cytokines as well as different hormonal imbalance by this bacterium, which are associated with diabetes mellitus. On the other hand, by tracing anti-H. pylori antibodies in patients with diabetes mellitus and occurrence of symptoms such as digestive problems in >75% of these patients, it can be concluded that there is a relationship between this bacterium and T2DM. Considering the evidence, it is crucially important that the probability of infection with H. pylori is evaluated in patients with T2DM so that medical process of the patient is followed with higher cautious.

Gamma Glutamyltransferase Levels and Its Association with High Sensitive C - reactive protein in Sudanese Patients with Type 2 Diabetes Mellitus

Article

Full-text available

Jan 2013

The aim of the study is to examine the relationship between Gamma-glutamyltransferase and the marker of inflammation, i.e., high sensitive C reactive protein, in type 2 diabetic patients in Sudan. Materials and Methods: A prospective, analytical, hospital based, case control study was carried out Fedail Hospital and Khartoum Teaching Hospital. 100 patients (50 males and 50 females) presenting with type 2 Diabetes Mellitus were enrolled in this study compared to and 50 healthy control subjects from May 2012 to May 2013 were included in the study. Fasting blood glucose, Serum gamma-glutamyltransferase and High sensitive C-reactive protein were measured. Results: A significant positive correlation between Gamma Glutamyltransferase and high sensitivity-C-reactive protein in patients with type 2 diabetes (r = 0.431, p <0.001). Conclusion: These results suggested that high sensitivity-C-reactive protein levels increase continuously across the fasting blood glucose spectrum starting from the lowest fasting blood glucose in both men and women.

Beta Cell Function, Insulin Resistance and Low Grade Systemic Inflammation in Myanmar Adults with Different Categories of Glucose Tolerance

Article

Full-text available

May 2013

Objective. The aim of the study was to evaluate the beta cell function, insulin sensitivity and low grade systemic inflammation in different categories of glucose tolerance in Myanmar. Methodology. A cross-sectional study was conducted on 202 Myanmar subjects of both sexes, aged between 45-65 years old. Fasting blood glucose, insulin, C-peptide and hs-CRP levels were measured. A 75g oral glucose tolerance test was performed. Insulin resistance and beta cell function were assessed by homeostasis-model-assessment (HOMA). Results. The subjects were categorized as normal glucose tolerance (NGT), impaired fasting glucose (IFG), impaired glucose tolerance (IGT) and diabetes mellitus (DM) according to WHO-2006 criteria. Fasting serum insulin, C-peptide and hs-CRP levels and insulin resistance index (HOMA-IR) progressively increased from NGT through prediabetes (IFG, IGT) to DM (p<0.01). Beta-cell function did not change significantly in any other group as compared to normal group. Conclusion. After multivariate analysis, increases in fasting C-peptide, hs-CRP and HOMA-IR index were significantly associated with diabetes. It was also found that insulin resistance was a predominant feature in deterioration of the glucose tolerance in Myanmar subjects.

Levels of Highly Sensitive C-reactive Protein and Gamma-Glutamyltransferase in Sudanese Patients with Metabolic Disorders

Article

Full-text available

Jan 2014

Abstract Objectives: The objective of this study is to evaluate the levels of gamma-glutamyltransferase and highly sensitive C-reactive protein in patients with metabolic disorders in Sudan. Material and method: A case control hospital based study was carried out at Atbara-River Nile State-Sudan. Total of 294 participants, 194 patients were with at least one of the metabolic disorders (Diabetic, Obese, or Hypertensive) were enrolled in this study together with 100 healthy control subjects during the period from October to November 2013. For all participants, body mass index (BMI), gamma-glutamyltransferase (GGT) and highly sensitive C-reactive protein (hs-CRP) were measured using standard methods. Result: A significant elevation of GGT levels was observed in all patients (diabetic, hypertensive, and obese). Levels of hs-CRP significantly increased in obese and diabetic. While in hypertensive patients insignificant elevation was observed. Conclusion: patients with metabolic disorders, included in this study, have significantly elevated GGT levels and hs-CRP except hypertensive.

Intakes of Whole and Refined Grains and Dietary Fibre In Relation to Plasma Inflammatory Protein Concentrations

Article

Rachel C Masters

Helicobacter pylori infection and diabetes: Is it a myth or fact?

Article

Full-text available

Apr 2014
WORLD J GASTROENTERO

Helicobacter pylori (H. pylori) is one of the most common human bacterial pathogens, and infection causes a wide array of gastric disorders, including simple gastritis, peptic ulcers and gastric malignancies. Gastrointestinal inflammation caused by H. pylori can influence the absorption of glucose and lipids, which are also abnormal in diabetes mellitus. Type 2 diabetes mellitus (T2DM), formerly known as non-insulin-dependent diabetes mellitus or adult-onset diabetes, is a metabolic disorder that is characterized by high levels of blood glucose resulting from insulin resistance and relative insulin deficiency. It is an emerging pandemic and is rapidly becoming a serious threat to public health. Emerging data now indicate a strong relationship between H. pylori infection and the incidence of T2DM. The mechanisms underlying the pathogenesis of diabetes are complex, involving insulin resistance, chronic inflammation, insulin secretion deficiency as a result of pancreas β-cell dysfunction, glucotoxicity, and lipotoxicity. H. pylori infection is known to be involved in the pathogenesis of insulin resistance, and the growing awareness of its role in diabetes is important for the early detection of glucose dysregulation and prevention of T2DM in high-risk communities. This review probes the possible relationship between H. pylori and diabetes according to epidemiological surveys and discusses putative mechanisms underlying this correlation.

The high risk for type 2 diabetes among ethnic minority populations is not explained by low-grade inflammation

Article

Full-text available

Dec 2019

Our aim was to identify whether low-grade inflammation, reflected by C-reactive protein (CRP), explains the higher risk for incident type 2 diabetes (T2D) among ethnic minorities. We included 837 Dutch, 712 South-Asian Surinamese, 797 African Surinamese, 804 Ghanaian, 817 Turkish and 778 Moroccan origin participants of the HELIUS study (Amsterdam, the Netherlands). We used multiple linear regression to assess ethnic differences in CRP levels. We determined the association of CRP with T2D and the modifying effect of ethnicity by cox regression, and compared hazard ratios for the association between ethnicity and T2D before and after adjustment for CRP. CRP levels were higher in ethnic minority groups than in Dutch origin participants. CRP was associated with a higher T2D incidence, similarly across ethnic groups (overall HR per SD 1.38 [95% CI 1.14; 1.68]). However, the association was attenuated and no longer statistically significant after adjustment for adiposity measures (HR 1.11 [95% CI 0.90; 1.37]). CRP accounted for a very small part of the ethnic differences in T2D, but only in models unadjusted for adiposity. Low-grade inflammation does not substantially contribute to the higher risk of T2D among ethnic minority populations compared to the Dutch.

2Adrenergic Receptor Polymorphism and Obesity in Type 2 Diabetes: Response to Yoshioka, Yoshida, and Yoshikawa

Article

May 2003
DIABETES CARE

We recently reported that our group ( n = 502) of type 2 diabetic subjects did not reveal statistically significant differences in BMI with respect to the three groups of carriers of β2-adrenergic receptor polymorphisms (1). Meirhaghe et al. (1) report that in men not participating in physical activity, Gln27Gln carriers had a higher BMI than the Glu27 (combined heterozygous and homozygous) carriers (27.2 ± 0.4 vs. 25.2 ± 0.3 kg/m2), while no effect on BMI was found …

Prospective Relation of C-Reactive Protein With Type 2 Diabetes: Response to Snijder et al

Article

Full-text available

May 2003
DIABETES CARE

Dietary factors and the risk of developing insulin dependent diabetes in childhood

Article

Full-text available

Jun 1990

To study different nutrients and food additives as risk factors for insulin dependent diabetes mellitus in childhood. Prospective case-control study. Parents of the children being studied were asked to fill in a questionnaire regarding the children's frequency of consumption of various foods. Parents of children with diabetes were asked about the period before onset of the disease. Population based study throughout Sweden. 339 Children aged 0-14 who had recently developed insulin dependent diabetes mellitus and 528 control children matched for age, sex, and county of residence who were traced through the official Swedish population register. Foods were classified according to their content of protein, fat, carbohydrates, monosaccharides or disaccharides, nitrosamines, nitrates or nitrites, vitamin C, and fibres. The frequency of intake was categorised as high, medium, and low and the relative risk for developing insulin dependent diabetes was estimated for the three frequencies of intake and calculated as odds ratios. Significant linear trends for dose response in odds ratios by frequency of intake were shown for solid foods containing high amounts of protein (odds ratio for low frequency of intake 1.0; medium 2.3; and high 5.5), and nitrosamines (1.0; 1.7; 2.6) and significant but non-linear trends were found for carbohydrates (1.0; 1.3; 4.4) and nitrates or nitrites (1.0; 0.8; 2.4). The significant trends were not affected when the results were standardised for possible confounders. No significant increases in odds ratios were found for protein, monosaccharides and disaccharides, vitamin C, and fibres. Nutrients and food additives such as protein, carbohydrate, and nitrosamine compounds may influence the risk of developing insulin dependent diabetes in childhood and significant trends in odds ratios indicate a causal relation.

Prevalence and Determinants of Glucose Intolerance in a Dutch Caucasian Population: The Hoorn Study

Article

Full-text available

Sep 1995
DIABETES CARE

To study the prevalence and determinants of glucose intolerance in a general Caucasian population. A random sample of 50- to 74-year old Caucasians (n = 2,484) underwent oral glucose tolerance tests. Multiple regression analyses were performed to study the association of 2-h postload plasma glucose values with potential determinants. Prevalence of known and newly detected diabetes and impaired glucose tolerance was 3.6, 4.8, and 10.3%, respectively. In women, but not in men, the association of body mass index with 2-h glucose was fully accounted for by the waist-to-hip ratio. Maternal history of diabetes was twice as prevalent as paternal history, but paternal history only was associated with 2-h glucose. In addition, paternal history was a stronger determinant in men than in women. An independent positive association with 2-h plasma glucose was found for alcohol use of > 30 g/day in women and for intake of total protein, animal protein, and polyunsaturated fatty acids in men. An independent inverse association with 2-h plasma glucose was demonstrated for height (both sexes), alcohol use of < or = 30 g/day (both sexes), energy intake (in men), and, unexpectedly, current smoking (in men). The prevalence of diabetes in elderly Caucasians was 8.3%. In men, dietary habits may unfavorably influence glucose tolerance independent of obesity.

??-Glutamyl transpeptidase-dependent iron reduction and LDL oxidation - A potential mechanism in atherosclerosis

Article

Full-text available

Apr 1999

gamma-Glutamyl transpeptidase (gamma-GT) is found in serum and in the plasma membranes of virtually all cell types. Its physiologic role is to initiate the hydrolysis of extracellular glutathione (GSH), a tripeptide in which cysteine lies between alpha-glycine and gamma-glutamate residues. Cysteine and other thiol compounds are known to promote LDL oxidation by reducing Fe(III) to redox active Fe(II); therefore, we sought to determine whether similar reactions can be sustained by GSH and influenced by gamma-GT. Fe(III) reduction and LDL oxidation were studied by monitoring the formation bathophenanthroline-chelatable Fe(II) and the accumulation of thiobarbituric acid-reactive substances, respectively. Human atheromatous tissues were examined by histochemical techniques for the presence of oxidized LDL and their colocalization with cells expressing gamma-GT activity. A series of experiments showed that the gamma-glutamate residue of GSH affected interactions of the juxtaposed cysteine thiol with iron, precluding Fe(III) reduction and hence LDL oxidation. Both processes increased remarkably after addition of purified gamma-GT, which acts by removing the gamma-glutamate residue. GSH-dependent LDL oxidation was similarly promoted by gamma-GT associated with the plasma membrane of human monoblastoid cells, and this process required iron traces that can be found in advanced or late stage atheromas. Collectively, these findings suggested a possible role for gamma-GT in the cellular processes of LDL oxidation and atherogenesis. Histochemical analyses confirmed that this may be the case, showing that gamma-GT activity is expressed by macrophage-derived foam cells within human atheromas, and that these cells colocalize with oxidized LDL. Biochemical and histochemical correlates indicate that gamma-GT can promote LDL oxidation by hydrolyzing GSH into more potent iron reductants. These findings may provide mechanistic clues to the epidemiologic evidence for a possible correlation between persistent elevation of gamma-GT and the risk of fatal reinfarction in patients with ischemic heart disease.

Reduced insulin-like growth factor-1 associated with insulin resistance in syndrome X patients

Article

Aug 2000

Therapeutic Angiogenesis in Critical Limb and Myocardial Ischemia

Article

Oct 2001
J Intervent Cardiol

Research in animal models of ischemia has shown that administration of angiogenic growth factors, either as a recombinant protein or by gene transfer, can augment nutrient perfusion through neovascularization to promote the development of supplemental collateral blood vessels that will constitute endogenous bypass conduits around occluded native arteries; a strategy termed “therapeutic angiogenesis.” In animal models and clinicul trials, the best studied cytokines with angiogenic activity are vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF). Clinical trials of therapeutic angiogenesis in patients with critical limb ischemia demonstrated resolution of rest pain and/or improved limb integrity, increased pain-free walking time and ankle-brachial index, newly visible collateral vessels by digital subtraction angiography, and qualitative evidence of improved distal flow by magnetic resonance imaging. Initial clinical trials in patients with end-stage coronary artery disease using direct myocardial injection via thoracotomy resulted in large increases in exercise time and marked reductions in anginal symptoms, as well as objective evidence of improved perfusion and leji ventricular function. Larger scale placebo-controlled trials have been limited to intracoronary and intravenous administration of recombinant protein, and have not shown significant improvement in exercise time or angina compared to placebo. Larger scale placebocontrolled studies of gene transfer using catheter-based endocardial delivery are in progress. Future clinical studies are required to determine the optimal dose, formulation, route of administration, and combinations of growth factors, as well as the requirement for endothelial progenitor cell or stem cell supplementation, to provide effective and safe therapeutic angiogenesis for patients with critical limb ischemia and chronic myocardial ischemia who are not candidates for conventional revascularization procedures.

Article

Sep 1992

Humoral immune factors related to type 1 diabetes have been investigated in children with coeliac disease. Anti-insulin (IAAb), immunoglobulin (alpha IgAb), islet cell (ICA) and glucagon autoantibodies were examined in 15 children with coeliac disease at diagnosis (group 1), in 15 children with coeliac disease following a gluten-free diet (group 2) and in 30 control patients (groups 3 and 4). IAAb were present in 27% of group 1 and in 20% of group 2 patients and alpha IgAb were significantly increased in group 1 and 2 patients; two patients in group 2 were positive for ICA; none of the coeliac disease patients were positive for anti-glucagon antibodies. The levels of anti-gliadin antibodies in group 1 were positively correlated with those of alpha IgAb. Coeliac disease-related HLA antigens were not correlated with antibody presence. The presence of diabetes-related humoral immune factors in coeliac disease raises the question as to whether or not they are predictive of subclinical pancreatic damage or whether they are simply indicators of a more general autoimmune diathesis.

Insulin growth factor 1 increases vascular smooth muscle nitric oxide production

Article

Feb 1997
LIFE SCI

Insulin like growth factor 1 (IGF-1) vasodilates, increases blood flow and lowers blood pressure; nitric oxide (NO) has been suggested to mediate some of these effects. We studied the role of IGF-1 in the regulation of NO production in vascular smooth muscle cells (VSMC). IGF-1 induced a concentration and time-dependent increase in NO release from endothelium-denuded aortic rings. Pre-incubation with cycloheximide or aminoguanidine blocked IGF-1-stimulated NO release. In addition, a six-fold increase in NO production by VSMC was seen upon incubation with IGF-1. These results suggest that IGF-1 induces NO release in VSMC through a process that involves new protein synthesis and the inducible isoform of nitric oxide synthase.

??2-Adrenoceptor gene polymorphism, body weight, and physical activity

Article

Apr 1999

Insulin-like growth factor binding protein-1 in NIDDM: Relationship with the insulin resistance syndrome

Article

Feb 1999

In order to examine the role of insulin-like growth factors in the pathogenesis of accelerated macrovascular disease in noninsulin-dependent diabetes mellitus (NIDDM), we investigated the relationship between the insulin resistance syndrome and the IGF axis. Cross-sectional analysis of the relationship between insulin resistance syndrome variables and concentrations of IGF-1, IGF-2, IGFBP-1 and IGFBP-3 in 80 subjects with NIDDM. After correcting for age, sex and body mass index, concentrations of IGFBP-1, correlated with those of HDL-cholesterol (r = 0.40; P < 0.001), triglycerides (r = -0.24; P = 0.04), insulin (r = -0.39; P < 0.001), intact proinsulin (r = -0.32; P = 0.006), des 31,32 proinsulin (r = -0.40; P = 0.001), and with insulin sensitivity (r = 0.38; P = 0.001) and PAI-1 activity (r = -0.24; P = 0.05); IGF-1 levels only correlated with those of HDL-cholesterol (r = -0.33; P = 0.005), and this was not explained by IGFBP-1 or insulin sensitivity. With additional correction for insulin, concentrations of IGFBP-1 still correlated with HDL-cholesterol (r = 0.40; P < 0.001), but not those of triglycerides or PAI-1 activity. There were no significant relationships between levels of IGF-2 and any of the variables investigated, and IGFBP-3 levels only correlated with those of total cholesterol (r = 0.24, P = 0.04). In NIDDM, concentrations of IGFBP-1 are related to those of insulin, insulin sensitivity, serum lipoproteins and PAI-1 activity. The relationship between concentrations of IGFBP-1 and HDL-cholesterol is not explained by insulin. Concentrations of IGF-1 are linked to HDL-cholesterol, and this is not explained by levels of IGFBP-1. IGFBP-1 concentrations were related to PAI-1 activity, and this may be explained by insulin, which regulates the production of IGFBP-1 and PAI-1.

Circulating Igf-I And Its Protective Role In The Pathogenesis Of Diabetic Angiopathy

Article

Feb 2000

Poor glycaemic control in type 1 diabetes is associated with elevated serum IGFBP-1 levels and reduced rather than elevated serum IGF-I levels. Increasing age is accompanied by a further decrease in serum IGF-I levels as well as an increase in IGFBP-l levels in adult diabetic type 1 and type 2 subjects. This is especially observed in diabetic type 1 subjects with manifest microvascular complications. IGFBP-I has been proposed as one of the IGF-I inhibitors in the serum of diabetics. Lowered IGF-I and increased IGFBP-1 levels in the blood may thus result in decreased IGF-I bioavailability at the tissue level. We hypothesize that the premature and progressive decline in serum IGF-I bioactivity during ageing in diabetics ultimately results in insufficient protective effects by IGF-I in the kidneys, eyes and neurones, and thus the progression of diabetic microvascular complications. If this hypothesis is proven to be right, treatment of diabetic patients with IGF-I (eventually complexed to IGFBPs) as an adjunct to insulin might prevent and not worsen the development of diabetic microvascular complications.

Prospective Relation of C-Reactive Protein With Type 2 Diabetes: Response to Han et al

Figures

Recommended publications

Last Chance to Apply! 18 Assistant & Associate Professor positions at VU Amsterdam

See our 18 open Asst/Assoc. Professor positions at the science faculty of VU Amsterdam

Effect of enalapril on the incidence of retinopathy in normotensive patients with type 2 diabetes

A study evaluating prevalence of hypertension and risk factors affecting on blood pressure control a...

Prospective relation of C-reactive protein with type 2 diabetes

Effectiveness of the Austrian disease management programme "Therapie Aktiv" for type 2 diabetes rega...