Cerebrovascular Reactivity and ReserveCapacity in Type II Diabetes MellitusBela Fulesdi, Martien Limburg, Daniel Bereczki, Miklos Kaplar, Csilla Molnar, Janos Kappelmayer,Gyula Neuwirth, and Laszlo Csiba

ABSTRACT

The aim of the study was to test the hypothesis that laboratory parameters and cerebrovascularreserve were investigated by linear regressioncerebrovascular reserve capacity and cerebrovascularanalysis. Resting cerebral blood flow velocity wasreactivity are impaired in patients suffering fromsimilar in controls and in the two diabeticnon insulin-dependent diabetes mellitus. We alsosubgroups. Cerebrovascular reactivity wasintended to investigate factors which may influenceelevated for a shorter time in patients with . 10resting cerebral blood flow velocity andyears disease duration than in controls and short-cerebrovascular reserve capacity. A total of 28term diabetic patients. Cerebrovascular reservepatients suffering from type II diabetes mellituscapacity was lower in the long-term diabetesand 20 healthy control subjects were studied. Basedgroup (means 6 SD: 39.6 6 20.7%) than in patientson diabetes duration patients were divided intowith < 10 years disease duration (63.3 6 17.4%,two groups: subjects with . 10 years and thosep , 0.02 after Bonferroni correction).with < 10 years disease duration. Middle cerebralCerebrovascular reserve capacity was inversely

artery mean blood flow velocities were measured related to the duration of the disease (R 5 0.53,at rest and after intravenous administration of 1g p , 0.003). None of the determined laboratoryacetazolamide. Cerebrovascular reactivity and factors had any relation with resting cerebralreserve capacity were calculated. Blood glucose, blood flow and cerebrovascular reserve capacity.insulin, glycosylated hemoglobin, hemostatic The vasodilatory ability of cerebral arterioles isfactors (fibrinogen, alpha-2 macroglobulin and von diminished in long-standing type II diabetesWillebrand factor antigen) were determined. mellitus. (Journal of Diabetes and ItsCerebrovascular reactivity and reserve capacity Complications 13; 4: 191–199, 1999.) 1999 Elseviervalues were compared between the two diabetic Science Inc.subgroups and controls. Correlations between

INTRODUCTION

Diabetes mellitus is one of the most commondiseases in the industrialized countries.

Department of Neurology (B.F., D.B., L.C.), 2nd Department ofCardiovascular and cerebrovascular com-Internal Medicine Neurosurgery and Clinical Chemistry University

Medical School of Debrecen, 1st Department of Internal Medicine plications, nephropathy and neuropathyKenezy County Hospital, Debrecen, Hungary, and Departments of are the main secondary affections. In diabetic individu-Neurology (B.F., M.L.) and Clinical Informatics (M.L.), Academic

als cardiovascular and cerebrovascular complicationsMedical Center Amsterdam, The Netherlands.B.F. worked as a research fellow at the AMC, Amsterdam. M.L. are responsible for the majority of fatal outcomes.1–8

is a Clinical Investigator of the Netherlands Heart Foundation. Therefore, adequate glycemic control, widely usedReprint requests to be sent to: Dr. Bela Fulesdi, Department of

screening and preventive treatment of vascular compli-Neurology, University Medical School of Debrecen, Nagyerdei krt.98. Debrecen, Hungary, H-4012. cations are essential in order to reduce morbidity and

Journal of Diabetes and Its Complications 1999; 13:191–199 1999 Elsevier Science Inc. All rights reserved. 1056-8727/99/$–see front matter655 Avenue of the Americas, New York, NY 10010 PII S1056-8727(99)00044-6

192 FULESDI ET AL. J Diab Comp 1999; 13:191–199

mortality.9–12 However, prevention and screening of blood flow velocity and cerebrovascular reserve ca-pacity in patients suffering from type II diabetescerebrovascular complications remain a stepchild of

routine diabetes care. mellitus?In the past decade noninvasive ultrasound tech-

METHODSniques have gained increasing importance for cerebro-

Between April 1994 and May 1997 outpatients withvascular diagnosis and screening. High resolution du-type II (non–insulin-dependent) diabetes mellitus fromplex scanning of the carotid arteries enables accuratethe diabetes outpatient clinics of the 2nd Departmentassessment of intima-media ratio of the carotid wall,13

of Internal Medicine, Debrecen and the Academic Med-accurate measurement of the diameter of carotid arter-ical Center, Amsterdam entered the study. The inclu-ies and examination of plaque surface.14 However, upsion criteria were: age over 18 years and non–insulin-to now no routinely used, non-invasive diagnostic test-dependent diabetes mellitus. The patients had to being method is available for assessing the function ofnormotensive and without cerebrovascular disease incerebral microvessels.the history. Hemodynamically significant stenoses ofAssessment of cerebral microvascular function isthe common and internal carotid arteries were ex-more complicated, as cerebral microvessels are locatedcluded with duplex scanning (Hewlett-Packard Sonoswithin the bony skull, which is relatively impermeable2000, USA). Those with hemodynamically significantto ultrasound. Frequently used methods for this pur-stenosis did not enter the study.pose are measuring cerebral blood flow with single

Blood was sampled for determination of the actualphoton emission computed tomography (SPECT)15,16 orblood glucose and insulin, glycosilated hemoglobinmeasuring cerebral blood flow velocity in the basal(HbA1C) and the following hemostatic factors: fibrino-cerebral arteries using transcranial Doppler (TCD)17–19

gen, von Willebrand factor antigen and alpha-2 macro-at rest, or after administration of a vasodilatory stimu-globulin levels. Blood glucose was determined by thelus, e.g., CO2-inhalation20 and acetazolamide,21 andglucose oxydase-peroxydase method (Hitachi 717, Ky-performing repeated measurements thereafter. Vaso-oto, Japan). Insulin was determined by 125I-insulin ra-dilatory ability of the cerebral microvasculature can bedioimmunassay (JNG 403 RIA Multidetector Counter,calculated based on the increase of cerebral blood flowVILLA, Spisska Nova Ves, Slovakia). Hemoglobin A1C(SPECT, PET) or cerebral blood flow velocity (TCD).was assessed by DIAMAT (Biorad, Hercules, Califor-In recent years, these methods have indicated that he-nia, USA) equipment. Fibrinogen measurements weremodynamically significant stenoses of the brain sup-performed with STA Compact Coagulometer (Diag-plying arteries17,20 and diseases affecting the brain mi-nostica Stago, Asnieres, France) by using the Clausscrovasculature (e.g. hypertension, diabetes mellitus,method. The von Willebrand factor antigen levels weresystemic lupus erythematosus) may cause a less inten-essayed by the STA Compact Coagulometer using im-sive vasomotor reactivity response, referring to a de-munoturbidimetric assay. Alpha-2 macroglobulin val-creased vasodilatory ability of cerebral arterioles.21–23

ues were measured by the Laurell rocket immunoelec-In previous studies it was demonstrated that vasodi-trophoresis technique.latory function of the cerebral arterioles is diminished

Resting middle cerebral artery mean blood flow ve-in patients with type I diabetes.21,24–28 In our previouslocity was measured on both sides in supine position atstudy21 we have shown that impairment of vasomotor50-mm depth using EME TC 64-B transcranial Dopplerreactivity is proportional to disease duration in insulin-(Eden Medizinische Elektronik, Uberlingen, Ger-dependent diabetic patients and correlates with othermany). Measurements were repeated at 5, 10, 15, anddiabetic complications, such as nephropathy and reti-20 min after intravenous injection of 1 g acetazolamidenopathy. Thus, cerebral microangiopathy in insulin-(Diamox, Lederle Parenterals, Puerto Rico, USA).dependent diabetes mellitus can be demonstrated andTranscranial Doppler measurements were performedused for research. No previous work has been pub-on both MCAs. Cerebrovascular reactivity (percent in-lished so far, investigating acetazolamide-inducedcrease of middle cerebral blood flow velocity after acet-vasomotor-reactivity responses in type II (non–insulin-azolamide at any time point of the measurement) anddependent) diabetes mellitus. Therefore, the aim of thecerebrovascular reserve capacity (maximal percent in-present study was to answer the following questions:crease of middle cerebral artery mean blood flow veloc-

1. Is resting cerebral blood flow velocity in patients ity after acetazolamide) were calculated as describedsuffering from type II diabetes mellitus different earlier.21

from healthy control subjects? Based on disease duration patients were arbitrarily2. Is there any difference in cerebrovascular reactivity divided into two groups: long-term patients with more

and reserve capacity between healthy persons and than 10 years and short-term patients with 10 years ordiabetic individuals? less disease duration.

Data of cerebrovascular reactivity and reserve capac-3. What are factors that influence resting cerebral

J Diab Comp 1999; 13:191–199 CEREBRAL VASOREACTIVITY IN TYPE II DIABETES MELLITUS 193

TABLE 1. DATA OF 28 INVESTIGATED SUBJECTS WITH DIABETES MELLITUS TYPE II AND THEREFERENCE VALUES

Value 6 SD inParameter Diabetic Patients Reference Values

Age (years) 57.1 6 12.1 —Diabetes duration (years) 10.4 6 6.5 —MCA rest (cm/sec) 57.6 6 15.6 —Cerebrovascular reserve (%) 54.1 6 19.8 —Glucose (mmol/L) 10.8 6 3.9 3.6–6.0Insulin (mU/L) 27.8 6 18.7 5–25Fibrinogen (g/L) 3.9 6 0.9 1.5–4.0von Willebrand factor antigen (%) 132.2 6 57.6 60–150Alpha-2 macroglobulin (%) 128.9 6 64 102.6 6 14.8Microalbuminuria (mg/day) 32.5 6 37.5 ,30Hb A1C (%) 9.72 6 3.7 4.2–6.1

ity in patients were compared with 20, age- and gen- was performed to examine the relationship betweendisease duration of diabetes and cerebrovascular re-der-matched healthy volunteers (friends, colleagues,

relatives of physicians, nurses). All subjects gave in- serve capacity and resting cerebral blood flow velocity.Similar to this, correlations between hemostatic vari-formed consent. The study was approved by the Local

Ethical Committees of the University Medical School ables, insulin and glucose content and cerebrovascularreserve capacity and resting cerebral blood flow veloc-of Debrecen and Academic Medical Center of Am-

sterdam. ity were assessed by linear regression analysis. Statis-tica for Windows version 5.1 (Statsoft Inc. Tulsa, USA)Statistical analysis: Means and standards deviations

are reported for all values. Variables with normal dis- software was used for statistical analysis.tribution were compared with appropriate unpaired

RESULTSt-tests. We used repeated measure analysis of varianceto detect differences in mean MCA blood flow velocity Twenty-eight patients suffering from type II diabetes

mellitus, 14 females and 14 males were studied. Theand cerebrovascular reactivity after acetazolamide ad-ministration between the two diabetic and control main characteristics of the subjects are summarized in

Table 1. The mean 6 SD values for diabetes durationgroups. Pairwise comparisons were performed be-tween the groups for each timepoint by appropriate in the different subgroups were: 6.38 6 2.8 years in

short-term and 17.7 6 4.8 years in long-term diabetict-testing using Bonferroni correction for multiple com-parisons, when significant differences were detected patients.among the three groups. Analysis of variance (AN- Middle Cerebral Artery Mean Blood Flow VelocityOVA) was used to compare cerebrovascular reserve Measurements. Middle cerebral artery blood flowcapacity among the three groups. The Tukey test was velocities (MCAV) of the different diabetic subgroupsused for post hoc analysis. Linear regression analysis at rest and after acetazolamide stimulation are summa-

rized in Table 2. Middle cerebral artery mean bloodflow velocity (MCAV) was similar in all groups at rest.

TABLE 2. ABSOLUTE MEAN BLOOD FLOW No significant group main effect could be detected byVELOCITIES (6SD) IN THE MIDDLE CEREBRAL statistical testing of the MCAVs of the different groups

ARTERY AT REST AND AT DIFFERENT TIME (P 5 0.923), suggesting that mean blood flow velocitiesINTERVALS AFTER ADMINISTRATION OFwere similar in all groups after administration of aceta-ACETAZOLAMIDE IN THEzolamide. There was no significant group-time interac-DIFFERENT SUBGROUPStion (P 5 0.054), i.e., the time course of absolute blood

Time after Diabetes Diabetes flow velocity in the MCA did not differ among theAcetazolamide Controls #10 years .10 yearsgroups. Thus, absolute velocities at rest and after aceta-(min) (n 5 20) (n 5 16) (n 5 12)zolamide were similar in healthy controls and in the

0 56.4 6 15.1 56.0 6 11.6 61.0 6 20.8 diabetic subgroups. The reactivity curves had a similar5 74.8 6 20.5 78.9 6 18.1 74.3 6 30.1 fashion in all investigated groups. Administration of

10 80.8 6 21.9 83.7 6 23.2 83.6 6 25.2 acetazolamide resulted in a significant increase of the15 83.2 6 22.1 87.0 6 22.6 79.8 6 23.2 middle cerebral artery mean blood flow velocities in20 83.2 6 21.6 84.3 6 24.0 77.5 6 24.0 all groups as compared to resting values.

194 FULESDI ET AL. J Diab Comp 1999; 13:191–199

FIGURE 1 Cerebrovascular reactivity values in controls and in diabetic patients with 10 or fewer years and with greater than 10years disease duration. Values are depicted as means and SE.

Cerebrovascular Reactivity Calculations. The per- statistical significance. Thus, in general vasodilation islesser and lasts shorter time in patients with diabetescent increases of the middle cerebral artery mean bloodof greater than 10 years disease duration, suggestingflow velocity after acetazolamide stimulation (5 cereb-a decreased vasodilatory ability.rovascular reactivity, CVR) of the different groups are

depicted in Figure 1. A significant group main effect Cerebrovascular Reserve Capacity Calculations. Thewas present in the reactivity values of the three groups maximal percent increase of the middle cerebral artery(P 5 0.014), suggesting that cerebrovascular reactivity mean blood flow velocity (5 cerebrovascular reservevalues are different in the groups. From Figure 1, it is capacity, CRC) differs among the 3 groups (p 5 0.022).obvious that percent increase of the middle cerebral This difference is due to the significantly lower CRCartery mean blood flow velocity in long-term diabetic in patients suffering from diabetes for more than 10patients is less intensive after 15 and 20 min than in years as compared to patients with 10 years diseasecontrols and in short-term diabetic patients. Although duration or less (Tukey-test: p 5 0.017). The differencescerebrovascular reactivity in long-term diabetic pa- in CRC did not reach the level of statistical significancetients is also lower at 5 and 10 min than in controls between healthy controls and any of the diabetic sub-

groups (Figure 2).and short-term diabetes, these differences did not reach

J Diab Comp 1999; 13:191–199 CEREBRAL VASOREACTIVITY IN TYPE II DIABETES MELLITUS 195

FIGURE 2 Cerebrovascular reserve capacity values in controls and in diabetic patients with 10 or fewer years and with greater than10 years disease duration. Values are depicted as means and SE.

The Impact of Different factors on Resting Cerebral holding in patients with type II diabetes mellitus. TheyBlood Flow Velocity and Cerebrovascular Reactivity. found that diabetic patients with proliferative retinop-Based on linear regression analysis diabetes duration athy showed diminished vasodilatory responses aswas inversely related to cerebrovascular reserve capac- compared to those without retinopathy and with back-ity in type II diabetes mellitus (Figure 3). Neither rest- ground retinopathy. Previous studies in type I patientsing cerebral blood flow, nor cerebrovascular reserve resulted in similar findings.21,24–28 Vasodilatory abilitycapacity had a significant correlation with age of the of brain arterioles decreased with increasing diseasesubjects. Also, resting cerebral blood flow velocity and duration in type I patients and developed parallel withcerebrovascular reserve capacity were independent other diabetic complications (nephro- and retinopathy).21

from actual glucose and insulin levels, hemostatic vari- Among the factors we studied only duration of dia-ables (fibrinogen, von Willebrand factor antigen, betes had a significant impact on cerebrovascular re-alpha-2 macroglobulin), glycosylated hemoglobin lev- serve capacity. Although disease duration is a conti-els and microalbuminuria (Table 3). nous variable, for sake of clarity we decided to

dichotomize. We performed our analysis on a groupDISCUSSION of patients with a long disease duration (arbitrarily

defined as more than 10 years) and contrasted themWe demonstrated that cerebrovascular reactivity andwith a group of diabetic patients with a shorter diseasecerebrovascular reserve capacity are less intensive induration (arbitrarily defined as less than 10 years), aspatients with long-lasting type II diabetes mellitus andwe have done in a previous study in patients with typethat this impairment of vasodilatory ability is inverselyI diabetes mellitus.21 This has been decided based uponrelated to the duration of diabetes. To our best knowl-the observation that morphological changes of the mi-edge, this is the first study which investigated cerebralcrovessels develop approximately after 5–10 years dis-vasodilatory responses in type II diabetes mellitus afterease duration.41 Glucose and insulin-levels were notadministration of acetazolamide.correlated with resting cerebral blood velocity andIn a recently published study, Ceravolo et al.29 mea-

sured vasoreactivity of the brain arterioles after breath cerebrovascular reserve capacity in our sample. There

196 FULESDI ET AL. J Diab Comp 1999; 13:191–199

FIGURE 3 Correlation between cerebrovascular reserve capacity and duration of diabetes.

are no exact data on impact of actual glucose level on tected in insulin-dependent diabetic subjects, whichwas not related to blood glucose concentration.32 Inresting cerebral blood flow in type II diabetes mellitus.

The majority of these studies has been performed in children with diabetic ketoacidosis, cerebral blood flowwas increased, but this increase may be explained bytype I patients. Cerebral blood flow was increased in

diabetic patients with poor glycemic control.30,31 In other the decrease of the extracellular pH and not the bloodglucose concentration.33 The role of insulin in determin-studies an “instability” of cerebral blood flow was de-

TABLE 3. CORRELATION BETWEEN RESTING CEREBRAL BLOOD FLOW VELOCITY (MCAV),CEREBROVASCULAR RESERVE CAPACITY (CRC) AND THE MEASURED LABORATORY

VARIABLES IN DIABETIC INDIVIDUALS

MCA (cm/sec) CRC (%)

Factor r P r P

Age (years) 0.14 0.34 0.14 0.49Disease duration (years) 0.05 0.78 0.53 0.003Glucose (mmol/L) 0.31 0.12 0.20 0.31Insulin (mU/L) 0.10 0.68 0.08 0.97Fibrinogen (g/L) 0.09 0.65 0.31 0.13von Willebrand antigen (%) 0.07 0.75 0.35 0.14alpha-2 macro-globulin (%) 0.12 0.60 0.02 0.92Microalbumin uria (mg/day) 0.05 0.77 0.07 0.71HbA1c (%) 0.24 0.21 0.35 0.06

r 5 correlation coefficient, P 5 level of significance.

J Diab Comp 1999; 13:191–199 CEREBRAL VASOREACTIVITY IN TYPE II DIABETES MELLITUS 197

ing cerebral blood flow and its vasoreactivity is also tes,21 it is obvious that although a reduced cerebralcontroversial. In non–insulin-dependent diabetes mel- vasodilatory ability was observed both in type I andlitus elevated levels were related to cerebral microvas- type II diabetes, the mechanisms for impaired cerebralcular disease.34 Even in nondiabetic individuals, hyper- microvascular function are less clear in type II diabetesinsulinemia is an independent contributor to carotid mellitus. As non–insulin-dependent diabetes mellitusatherosclerosis.35 Hyperinsulinemia and insulin resis- is a more complex pathophysiological process, not onlytance could be crucial for the development of athero- diabetes itself, but also hypertension and hypercholes-sclerosis. Hyperinsulinemia increases intracellular Ca- terolemia should be taken into account. We recentlyconcentration,36,37 associated with increased atherogenic reported an impaired cerebrovascular reserve in hy-activity in the vessel wall.38 Although higher insulin lev- pertension.22 Thus, cerebral vasodilatory responses inels increase nitric oxide secretion of the endothelium,39 type II diabetes could be also modified by accompa-in insulin resistance the counteracting effect of nitric nying hypertension.oxide on the atherosclerotic process is missing. Serum What are the clinical implications of the results?level of insulin did not influence resting cerebral blood Based on our study vasodilatory ability of the brainflow velocity and cerebrovascular reserve capacity in resistance arterioles decreases with increasing durationour sample. Although higher serum insulin levels may of diabetes, pointing to the presence of cerebral mi-lead to an increased atherosclerotic activity in cerebral croangiopathy. This finding may be causally relatedvessels, the correlation between higher serum insulin to vascular dementia which is observed more fre-concentrations and vascular damage is not linear. Mul- quently in diabetic individuals,44 and the higher com-tiple insulin measurements will increase representivity. plication rates after major hypotensive anesthesia.25 De-

Increased levels of some hemostatic factors have creased microvascular reactivity may be responsiblebeen reported in patients with diabetes mellitus.40–43

for a worse outcome of strokes in diabetic patients.8

Fibrinogen is an important risk factor for vascular dam- Prospective studies should prove the clinical signifi-age and an increased level of von Willebrand factor cance of impaired cerebrovascular reserve in type IIantigen is an indicator of endothelial damage. The level diabetes.of von Willebrand antigen is increased in diabetes mel-

ACKNOWLEDGMENTlitus and this is independent of previous glycemic con-trol.42 Higher levels of fibrinogen were present in pa- The work was supported by the grant of the Hungariantients with diabetes mellitus type I and in type II.40 In Academy of Sciences (Zsigmond Diabetes Grant). The au-our sample no correlation could be detected between thors thank Viola Molnar and Etelka Virag for excellent tech-

nical assistance during the study.fibrinogen, von Willebrand factor antigen, alpha-2macroglobulin and resting cerebral blood flow velocity

REFERENCESand cerebrovascular reserve capacity. In a further anal-ysis no difference could be detected for these factors 1. Asplund K, Hagg E, Helmers C, Lithner F, Strand T,

Wester PO: The natural history of stroke in diabeticbetween long-term and short-term diabetic patientspatients. Acta Med Scand 207:417–424, 1980.(means 6 SDs in diabetes with . 10 years and < 10

years duration: fibrinogen: 4.2 6 0.9 and 3.85 6 0.9 g/ 2. Burchfiel CM, Curb JD, Rodriguez BL, Abbott RD, ChiuD, Yano K: Glucose intolerance and 22-year stroke inci-L, VIII. factor-related antigen: 124.2 6 41.5 and 146 6dence. Stroke 25:951–957, 1994.66.6%, alpha-2 macroglobulin: 119.2 6 36.4 and 138.4 6

83.4% respectively). The fact that hemostatic variables 3. Currie CJ, Morgan CL, Gill L, Stott NCH, Peters JR:Epidemiology and costs of acute hospital care for cereb-are not associated with resting cerebral blood flowrovascular disease in diabetic and nondiabetic popula-and cerebral vasodilatory ability in diabetes, probablytions. Stroke 28:1142–1146, 1997.refers to an earlier development of diabetic hemostatic

alterations than cerebral microangiopathy. The same 4. Helgason CM: Blood glucose and stroke. Stroke 23:1–6,1998.was true for retino- and nephropathy and cerebrovas-

cular reserve capacity in our sample. All long-term 5. Kannel WB, D’Agostino RB, Wilson PWF, Belanger AJ,Gagnon DR: Diabetes, fibrinogen and risk of cardiova-diabetic subjects and all but four individuals in thesular disease. Am Heart J 120:672–680, 1990.short-term diabetes group showed diabetic retinopa-

thy (24 of 28 investigated patients). Thus, there was 6. Lai SM, Alter M, Friday G, Sobel E: A multifactorialanalysis of risk factors for recurrence of ischemic stroke.no difference in the incidence of retinopathy betweenStroke 25:958–962, 1994.the two subgroups. Similar to this, although microal-

buminuria was higher in the long-term diabetes group 7. Mast H, Thompson JLP, Lee SH, Mohr JP, Sacco R:Hypertension and diabetes mellitus as determinants ofas compared to short-term diabetes, a significant corre-multiple lacunar infarcts. Stroke 26:30–33, 1995.lation between cerebrovascular reserve and level of

microalbuminuria was not present. If one compares 8. McCall AL: The impact of diabetes on the CNS. Diabetes41:557–570, 1992.the present results with those observed in type I diabe-

198 FULESDI ET AL. J Diab Comp 1999; 13:191–199

9. Tuomilehto J, Rastenyte D, Jousilahti P, Sarti C, Varti- 26. Dandona P, James IM, Newbury PA, Woollard ML,Beckett AG: Cerebral blood flow in diabetes mellitus:ainen E: Diabetes mellitus as a risk factor for death from

stroke. Stroke 27:210–215, 1996. evidence of abnormal cerebrovascular reactivity. BMJ2:325–326, 1978.10. Weinberger J, Biscarra V, Weisberg MK, Jacobson JH:

27. De Chiara S, Mancini M, Vaccaro O, Riccardi G, FerraraFactors contributing to stroke in patients with athero-IA, Gallotta G, Cicerano U, Postiglione A: Cerebrovas-sclerotic disease of the great vessels: The role of diabetes.cular reactivity by transcranial Doppler ultrasonogra-Stroke 14:709–712, 1983.phy in insulin-dependent diabetic patients. Cerebrovasc11. Alter M, Lai SM, Friday G, Singh V, Kumar VM, Sobel Dis 3:111–115, 1993.E: Stroke recurrence in diabetics. Does control of blood

28. Rodriguez G, Nobili F, Celestino MA, Francione S, Gulliglucose reduce the risk? Stroke 28:1153–1157, 1997.G, Hassan K, Marenco S, Rosadini G, Cordera R: Re-12. Krans HMJ: The guidelines for the St. Vincent Declara- gional cerebral blood flow and cerebrovascular reactiv-tion. Diabetologia 13:13–15, 1993. ity in IDDM. Diabetes Care 16:462–469, 1993.

13. Niskanen L, Rauramaa R, Miettinen H, Haffner SM, 29. Cerevolo MG, Lagalla G, Lippera S, Gregorio F, Provin-Mercuri M, Uusitupa M: Carotid artery intima-media ciali: Proliferative retinopathy predicts impairment inthickness in elderly patients with NIDDM and in nondi- brain vasomotor reserve in type II diabetes. Cerebrovascabetic subjects. Stroke 27:1986–1992, 1996. Dis 7:332–338, 1997.

14. Widder B, Paulat K, Hackspacher J: Morphological char- 30. Grill V, Gutniak M, Bjorkman O, Lindquist M, Stone-acterisation of carotid artery stenoses by ultrasound Elander S, Seitz RJ, Blomqvist G, Reichard P, Widen L:duplex scanning. Ultrasound Med Biol 16:349–354, 1990. Cerebral blood flow and substrate utilisation in insulin-

treated diabetic subjects. Am J Physiol 258:813–820, 1990.15. Chollet F, Celsis P, Clanet M, Guiraud-Chaumeil B, Ras-col A, Marc-Vergnes JP: SPECT study of cerebral blood 31. Neil HA, Gale EA, Hamilton SJ, Lopez Espinoza I, Kauraflow reactivity after acetazolamide in patients with tran- R, McCarthy ST: Cerebral blood flow increases duringsient ischemic attacks. Stroke 20:458–464, 1989. insulin-induced hypoglycemia in type 1 (insulin-depen-

dent) diabetic patients and control subjects. Diabetologia16. Dahl A, Russell D, Nyberg-Hansen R, Rootwelt K, Mow-30:305–309, 1987.inckel P: Simultaneous assessment of vasoreactivity us-

ing transcranial Doppler ultrasound and cerebral blood 32. Dandona P, James IM, Woolard ML, Newbury P: Insta-flow in healthy subjects. J Cereb Blood Flow Metab 14:974– bility of cerebral blood flow in insulin-dependent dia-981, 1994. betics. Lancet 2:1203–1205, 1979.

17. Chimowitz MI, Furlan AJ, Jones SC, Sila CA, Lorig RL, 33. Hoffman WH, Pluta RM, Fisher AQ, Wagner MB, Ya-Paranandi L, Beck-GJ: Transcranial Doppler assessment novski JA: Transcranial Doppler ultrasound assessmentof cerebral perfusion reserve in patients with carotid of intracranial hemodynamics in children with diabeticocclusive disease and no evidence of cerebral infarction. ketoacidosis. J Clin Ultrasound 23:517–523, 1995.Neurology 43:353–357, 1993.

34. Zunker P, Schick A, Buschmann HS, Georgiadis D, Na-18. Aaslid R, Markwalder TM, Nornes H: Noninvasive bavi DG, Edelmann M, Ringelstein EB: Hyperinsulinism

transcranial Doppler ultrasound recording of flow ve- and cerebral microangiopathy. Stroke 27:219–223, 1996.locity in basal cerebral arteries. J Neurosurg 57:769–774,

35. Bonora E, Willeit J, Kiechl S, Oberhollenzer F, Egger G,1982.Bonadonna R, Muggeo M: Relationship between insulin

19. American Academy of Neurology: Assessment: Trans- and carotid atherosclerosis in the general population.cranial Doppler. Neurology 40:680–681, 1990. Stroke 28:1147–1152, 1997.

20. Kleiser B, Widder B: Course of carotid artery occlusions 36. Draznin B, Sussman KE, Eckel RH, Kao M, Yost T,with impaired cerebrovascular reactivity Stroke 23:171– Sherman NA: Possible role of cytosolic free calcium174, 1992. concentrations in mediating insulin resistance of obesity

and hyperinsulinemia. J Clin Invest 82:1848–1852, 1988.21. Fulesdi B, Limburg M, Bereczki D, Michels RPJ, Neuw-irth Gy, Legemate D, Valikovics A, Csiba L: Impairment 37. Solini A, Di Virgilio F, Trevisan M, Cipollina MR, No-of cerebrovascular reactivity in long-term type I diabe- sadini R, Crepaldi G: Role of glucose and insulin ontes. Diabetes 46:1840–1845, 1997. calcium handling in type II diabetic patients with and

without hypertension and microalbuminuria [Ab-22. Ficzere A, Valikovics A, Fulesdi B, Juhasz A, Czurigastract]. Diabetologia 37(Suppl 1):A46, 1994.I, Csiba L: Cerebrovascular reactivity in hypertensive

patients. A transcranial Doppler study. J Clin Ultrasound 38. Phair RD: Cellular calcium and atherosclerosis: A brief25:383–389, 1997 review. Cell Calcium 9:275–284, 1988.

23. Csepany T, Valikovics A, Fulesdi B, Kiss E, Szegedi 39. Steinberg HO, Brechtel G, Johnson A, Fineberg N, BaronGy, Csiba L: Cerebral systemic lupus erythematosus AD: Insulin-mediated skeletal muscle vasodilation is[Letter]. Lancet 343:1103, 1994. nitric oxide dependent: a novel action of insulin to in-

crease nitric oxide release. J Clin Invest 94:1172–1179,24. Bentsen N, Larsen B, Lassen NA: Chronically impaired1994.autoregulation of cerebral blood flow in long-term dia-

betics. Stroke 6:497–502, 1975. 40. Kannel WB, McGee DL: Diabetes and cardiovasculardisease. JAMA 241:2035–2038, 1979.25. Croughwell N, Lyth M, Quill TJ, Newman M, Greeley

WJ, Smith LR, Reves JG: Diabetic patients have abnor- 41. Lorenzi M, Cagliero E: Pathobiology of endothelial andother vascular cells in diabetes mellitus. Diabetes 40:653–mal cerebral autoregulation during cardiopulmonary

bypass. Circulation 82(Suppl. 5):IV407–412, 1990. 659, 1991.

J Diab Comp 1999; 13:191–199 CEREBRAL VASOREACTIVITY IN TYPE II DIABETES MELLITUS 199

42. Myrup B, Rossing P, Jensen T, Gram J, Kluft C, Jesper- associated with diabetes and its complications. BMJ2:964–966, 1979.sen J: Procoagulant activity and intimal dysfunction in

IDDM. Diabetologia 38:73–78, 1995. 44. Schneider R, Kiecewetter H: The significance of micro-circulatory disturbances in the pathogenesis of vascular43. Fuller JH, Keen H, Jarrett RJ, Omer T, Meade TW, Chak-

rabarti R, North WRS, Stirling Y: Haemostatic variables dementia. Pharmacopsychiatry 21(Suppl 1):11–16, 1988.