microalbuminuria - diabetes care

MicroalbuminuriaImplications for micro- and macrovasculardisease

TORSTEN DECKERT, MD, DMSC

ALLAN KOFOED-ENEVOLDSEN, MD

KlRSTEN N0RGAARD, MD

KNUT BORCH-JOHNSEN, MD, DMSCBo FELDT-RASMUSSEN, MD, DMSCTONNY JENSEN, MD, DMSC

Microalbuminuria is diagnosed when the UAER is >20 but <200 |xg/min. Theprevalence of microalbuminuria among diabetic patients is 15-20%. Persistentmicroalbuminuria in diabetic patients is a risk marker not only of renal disease, butalso of proliferative retinopathy and cardiovascular morbidity and mortality. Evenamong nondiabetic individuals, those with microalbuminuria tend to have an in-creased cardiovascular morbidity. The established cardiovascular risk factors, such assmoking, elevated plasma cholesterol, fibrinogen, and hypertension, are seen morefrequently in diabetic patients with persistent microalbuminuria than in normoalbu-minuric diabetic patients of similar age, sex, and diabetes duration. However, theserisk factors cannot by themselves explain the cardiovascular overmortality in thesepatients. In addition, insulin resistance or genetic disposition to hypertension orcardiovascular disease fails to be the missing link. Accumulating evidence suggests acommon pathogenetic mechanism for microalbuminuria and premature atheroscle-rosis (i.e., qualitative alterations of the extracellular matrix, including decreaseddensity and sulfation of HS-PG). Decreased density of HS in the glomeruli may leadto albuminuria and mesangial proliferation. In the intima of large vessel walls,decreased density and/or sulfation of HS may enhance several of the processesinvolved in premature atherosclerosis. Diabetes affects the composition and structureof the extracellular matrix in many ways and leads to decreased density and sulfationof HS-PG by several mechanisms. Genetic differences in the sulfation of HS and/orgenetic defects in the coordinated biosynthesis of HS-PG might contribute to de-creased concentration and sulfation of HS-PG in susceptible individuals. It is hopedthat susceptibility genes can be identified soon, thereby making prevention of severelate diabetic complications more successful.

FROM THE STENO DIABETES CENTER, GENTOFTE, DENMARK.

ADDRESS CORRESPONDENCE AND REPRINT REQUESTS TO TORSTEN DECKERT, MD, STENO DIABETES

CENTER, 2, NIELS STEENSENS VEJ, D K - 2 8 2 0 GENTOFTE, DENMARK.

HS-PG, HEPARAN SULFATE-PROTEOGLYCAN; HS, HEPARAN SULFATE; IDDM, INSULIN-DEPENDENT DIA-

BETES MELL1TUS; ESRD, END-STAGE RENAL DISEASE; UAER, URINARY ALBUMIN EXCRETION RATE; N I D D M ,

NON-INSULIN-DEPENDENT DIABETES MELLITUS; G F R , GLOMERULAR FILTRATION RATE; G B M , GLOMERULAR

BASEMENT MEMBRANE; V L D L , VERY-LOW-DENSITY LIPOPROTE1N; T E R A , TRANSCAP1LLARY ESCAPE RATE OF

ALBUMIN; LDL, LOW-DENSITY LIPOPROTEIN; CVD, CARDIOVASCULAR DISEASE; ACE, ANGIOTENSIN-

CONVERTING ENZYME; I G G , 1MMUNOGLOBUL1N G ; S T Z , STREPTOZOC1N; C I , CONFIDENCE INTERVAL; E C G ,

ELECTROCARDIOGRAM.

A lbuminuria in IDDM is associatedwith an extremely high risk ofearly death. The relative mortality

(i.e., mortality in diabetic patients rela-tive to mortality in the sex- and age-matched background population) in dif-ferent age groups of patients withalbuminuria is —40 times higher than indiabetic patients without albuminuria(1). Patients with persistently Albustix-positive urine (Boehringer Mannheim,Indianapolis, IN), in whom other nondi-abetic renal diseases, urinary tract infec-tion, and cardiac insufficiency have beenexcluded, exhibit increasing blood pres-sure and declining GFR (2). These pa-tients suffer from clinical diabeticnephropathy. It is therefore tempting tobelieve that the difference in relativemortality between patients with andwithout albuminuria is attributable todeath from ESRD. However, this is notthe case.

ALBUMINURIA: A NEWCARDIOVASCULAR RISKMARKER— About 50% of the albu-minuric patients will die from cardiovas-cular causes before progressing to ESRD.Thus, after 25 yr of diabetes duration atage 40-45 yr, the cardiovascular mortal-ity (mostly from coronary heart diseaseor stroke) in albuminuric diabetics is 60times higher than in the backgroundpopulation, whereas in patients withoutalbuminuria, the relative cardiovascularmortality was only 2 -3 times that of thebackground population (3). Most ofthese patients die from coronary heartdisease. In a case control study of 59IDDM patients who were followed fromthe onset of albuminuria, coronary heartdisease developed 8 times more fre-quently than in a diabetic population ofsimilar age, sex, and diabetes duration(4) (Fig. 1). Albuminuria in IDDM is,therefore, not only a marker of renal dis-ease, but also a potent risk marker ofcardiovascular disorder.

Albuminuria is normally diag-nosed when strips such as Albustix are

DIABETES CARE, VOLUME 15, NUMBER 9, SEPTEMBER 1992 1181

Dow

nloaded from http://diabetesjournals.org/care/article-pdf/15/9/1181/440952/15-9-1181.pdf by guest on 27 January 2022

Microalbuminuria in micro- and macroxascular disease

WITH ALBUMINURIA (n = 59)

WITHOUT ALBUMINURIA (n = 59)

OF ALBUMINURIA

Figure 1—Cumulative incidence of coronary

heart disease in IDDM patients with and without

albuminuria. The two groups were carefully

matched regarding age, sex, and diabetes dura-

tion. All patients had normal ECG at the onset of

the study where UAER was 300 mg/24 h in the

albuminuric group. From Jensen et al. (4). © by

Diabetologia.

persistently positive. This corresponds toa UAER of >300 mg/24 h. However, innondiabetic patients and in young pa-tients with IDDM just after the onset ofthe disease, UAER is <30 mg/24 h (al-bumin/creatinin ratio in early morningurine <2.5 mg/mmol (5,6). Patients withclinical nephropathy, therefore, havepassed a stage where UAER is <300mg/24 h but >30 mg/24 h. In fact pro-spective studies using highly sensitivemethods for the determination of UAERhave demonstrated that patients who willlater develop clinical nephropathy arecharacterized by an exponential increaseof UAER (7,8). According to an interna-tional consensus conference, patientswith UAER between 30 and 300 mg/24h are, therefore, said to have microalbu-minuria, and patients with persistent mi-croalbuminuria are suffering from incip-ient nephropathy (9).

The prevalence of microalbumin-uria among diabetic patients is 15-20%(10,11). Persistent microalbuminuria inIDDM patients is associated with manyabnormalities (Table 1). Most of theseabnormalities will not be discussed inthis article in which we concentrate onthe implications of microalbuminuria onmacroangiopathy. The question here is:

Is microalbuminuria a risk marker notonly for clinical nephropathy (12), butalso for premature artherosclerosis? Thisis not known for IDDM patients of juve-nile onset, because the low cardiovascu-lar morbidity and mortality among theseyoung patients require huge populationstudies that have not yet been con-ducted. However, among NIDDM pa-tients, microalbuminuria is a potent car-diovascular risk marker, independent ofhypertension and hyperlipedimia (13).In these patients, albuminuria is muchmore strongly associated with prematuredeath from cardiovascular diseases thanwith ESRD. In a follow-up study of 175NIDDM patients with microalbuminuria(urinary albumin concentration > 15 mg/L), 120 (68%) patients died 10 yr afterthe onset of the study; but only 8 amongthese patients (7%) died from ESRD,whereas 58% died from acute myocar-dial infarction, cardiac failure, or stroke(13). Thus, microalbuminuria in NIDDMseems to be more relevant as a marker forCVD than for renal disease. Even in non-diabetic patients, microalbuminuria is arisk marker for premature atherosclero-

Prevalence of CVD in non-diabetics

N = Normoalbuminurla n = 114

M = Microalbuminuria n = 12

2p = 0.016

nFigure 2—Six years cumulative incidence of

CVD in 126 elderly nondiabetic patients with

normal UAER (<15 [ig/min; N) or microalbu-

minuria (M). Calculated from Damsgaard et al.

(15). © by the British Medical Journal.

sis, as demonstrated by Yudkin et al. (14)and confirmed by other groups (15).Thus, in a population-based study ofnondiabetic patients (normal fastingblood glucose), persons with microalbu-minuria (>15 (xg/min) also had a sever-al-fold increase in cardiovascular mortal-ity during the following 6 yr comparedwith persons with normal UAER (15)(Fig. 2). Thus, persistent albuminuria is a

Table 1—Nonrenal changes in IDDM patients with persistent microalbuminuria incomparison with patients having normal UAER

CHARACTERISTIC CHANGE

PROUFERAT1VE RETINOPATHY

BLOOD PRESSURE

CARDIAC OUTPUT

END DIASTOLIC VOLUME

WORKING CAPACITY

TERAALBUMIN CATABOLISM

PLASMA FIBRINOGEN

T E R FIBRINOGEN

FIBRINOGEN SYNTHESIS

FIBRINOGEN CATABOLISM

TOTAL CHOLESTEROL

VLDL

LDLVON WLLLEBRAND FACTOR

TISSUE PLASMINOGEN RESPONSE TO EXERCISE

ACE

MORE FREQUENT (76)

HIGHER (27)

DECREASED (77)

DECREASED (77)

DECREASED (78)

INCREASED (49)

INCREASED (79)

INCREASED (26)

INCREASED (BENT-HANSEN, UNPUBLISHED

OBSERVATIONS)

INCREASED (79)

INCREASED (79)

INCREASED (26)

INCREASED (26)

INCREASED (26)

INCREASED (47)

IMPAIRED (80)

INCREASED (81)

1182 DIABETES CARE, VOLUME 15, NUMBER 9, SEPTEMBER 1992

Dow


Deckert and Associates

Table 2—Clinical data, prevalence of death, cardiovascular risk factors and CVD inparents of type I (insulin-dependent) diabetic patients with nephropathy and in parents oftype I diabetic patients with normoalbuminuria (17)

SEX (F/M)AGE (YR)

PREVALENCE OF DEATH (%)

PREVALENCE OF CVD (%)

UAER (M-G/MIN)

PREVALENCE OF MICROALBUMINURIA (%)

SERUM CHOLESTEROL (MM)

PERCENTAGE OF SMOKERS (%)

PARENTS OF

WITH NEPHROPATHY

35/4158 ± 8 (43-79)10 (4-18)12 (6-21)4(1-214)5 (1-14)

5.7 ± 1.240 (29-52)

DIABETIC PATIENTS

WITH NORMOAL-

BUMINURIA

32/4558 ± 7 (43-71)8 (3-16)

13 (6-23)5 (1-68)

11 (4-21)5.8 ± 1.034 (24-46)

Values are means ± SD (range) or median (range) or prevalence (95%).

new potent cardiovascular risk markernot only in diabetics, but also in nondi-abetic "healthy" subjects.

GENETIC SUSCEPTIBILITY TOCARDIOVASCULAR DISEASE INIDDM PATIENTS WITHALBUMINURIA— Why is albumin-uria a marker of cardiovascular diseaseand mortality? Some observations sug-gested that diabetic patients with albu-minuria should have a genetic disposi-tion for CVD (16). Such observationswere not confirmed at our clinic (17).Thus, 153 nondiabetic parents of pa-tients with IDDM were examined for car-diovascular morbidity and mortality. Ofthose, 77 were parents to patients withnormal UAER, and 76 were parents topatients with nephropathy (Table 2). Nosignificant differences in number of car-diovascular deaths, prevalence of cardio-vascular morbidity, or frequency of car-diovascular risk factors were seenbetween these two groups (17). Otherobservations suggested that diabetic pa-tients with albuminuria should have agenetic disposition to hypertension(18,19). However, our own and othermore comprehensive studies found thatblood pressure was not different in par-ents of diabetic patients with and with-out nephropathy (20,21). Further insight

into the predisposition to CVD in dia-betic subjects with albuminuria comesfrom studies of cation transport systems,such as Na+/Li+ countertransport inerythrocytes. Diabetic subjects with al-buminuria usually have higher Na+/Li+

countertransport activity in erythrocytesthan diabetic patients without albumin-uria (20,22). However, differences be-

i 0.7-oE 0.5-

0.3-

0.2

j 0.05-

Parents ofdiabeticswithnephropathy

Parents ofdiabeticswith normo-albuminuria

Oldercontrols

Figure 3—Na + /L i + countertransport in non-

diabetic parents of IDDM patients with and

without clinical nephropathy. A group of control

subjects of similar age but without diabetes in

their families also is indicated. No difference was

seen between the groups. From Jensen et al. (20).

® by Diabetologia.

tween parents of patients with and with-out albuminuria were not alwaysobserved (20,21) (Fig. 3), indicating thatenvironmental factors are more likely toexplain the differences in Na+/Li+ coun-tertransport between albuminuric andnonalbuminuric patients than geneticfactors (23). Other abnormalities in cat-ion transport systems have been reportedin fibroblasts from diabetic patients withnephropathy kept in tissue culture forseveral weeks (24). These abnormalitiesmight be genetic markers of cellular dys-function, reflecting common pathoge-netic mechanisms for renal and cardio-vascular diseases in diabetic subjects.These results have to be confirmed, butthey indicate that albuminuria and pre-mature atherosclerosis may have somegenetic susceptibility factors in common.The multifactorial pathogenesis of albu-minuria and atherosclerosis, however,makes an identification of such factorshighly complicated.

ALBUMINURIA ANDCARDIOVASCULAR RISKFACTORS— Other observations indi-cated the possibility that albuminuria is amarker of CVD because established car-diovascular risk factors (such as smoking,elevated plasma concentration of fibrino-gen and cholesterol, and hypertension) aremore prevalent among patients withnephropathy. Several authors (includingourselves ([26]) have found a higher prev-alence of heavy cigarette smokers (25), el-evated fibrinogen, increased total choles-terol (26), and higher blood pressure inpatients with albuminuria compared withpatients with normal UAER (27). How-ever, according to our calculations, a com-bination of elevated blood cholesterol (by25%), fibrinogen (by 20%), blood pressure(by 15%), and heavy cigarette smoking canexplain only a small part of the 10-foldhigher cardiovascular mortality in such agroup of patients. Other more importantfactors must contribute. Also in NIDDM,albuminuria is associated with increasedblood pressure (11,28) and hyperlipe-demia (29,30), and hemostatic parameters,


Dow


Microalbuminuria in micro- and macrovascular disease

Table 3—Insulin resistance in IDDM patients without and with early nephropathy

N O NEPHROPATHY

EARLY NEPHROPATHY

n = number of patients.

INSULIN DOSE

(IU/KG, MEDIAN AND RANGE)

N = 391

0.72 (0.14-1.83)0.70(0.11-0.69)

FASTING

(PM

FREE PLASMA INSULIN

, MEANS ± SD)

N = 74

5055

±25±24

such as von Willebrand factor, and fibrin-ogen might be elevated in patients withovert clinical nephropathy (31). However,as Schmitz and Ingerslev (31) pointed outon the basis of their comprehensive study,the increased mortality among NIDDM pa-tients with microalbuminuria could not beexplained by coexisting risk factors, suchas hypertension, hyperlipidemia, or hemo-static disorders. Thus, from epidemiologi-cal studies, we must conclude that eleva-tion of blood pressure, cholesterol, andfibrinogen contribute, but cannot by them-selves, explain why albuminuria is such apotent cardiovascular risk marker. Re-cently, insulin resistance has been ac-knowledged as a potential cardiovascularrisk factor. However, insulin resistance asreflected by the daily insulin requirementor the fasting free plasma insulin concen-tration seems not to be more pronouncedin IDDM patients with early nephropathythan those without (26) (Table 3).

POSSIBLY COMMONPATHOGENETIC MECHANISMFOR ALBUMINURIA ANDPREMATUREATHEROSCLEROSIS— What is themissing link? We suggest a commonpathogenetic mechanism of microalbu-minuria and premature atherosclerosis,because of the close coincidence of albu-minuria and atherosclerotic events andbecause of the similarity of structural andfunctional alterations of glomeruli andlarge vessel walls in patients with albu-minuria. Mesangial and arterial myome-dial cells are mesenchymal cells, withcontractile properties equal to their ph-agycytotic properties. In patients with al-

buminuria, enhanced mesangial andmyomedial proliferation takes place.Both these cells can synthesize collagenIV, fibronectin, laminin, and HS-PG.This also is reflected by the compositionof the extracellular matrix, which is quitesimilar. In patients with albuminuria, notonly is accumulation of extracellular ma-trix more pronounced than in patientswithout albuminuria (32), but changesin the quality of the extracellular matrixseems to be a characteristic feature (Ta-ble 4).

DECREASED DENSITY OF HS-PGIN EXTRACELLULAR MATRIX OFDIABETIC ANIMALS ANDHUMANS— Shimomura and Spiro(33) have shown the density of HS-PG tobe reduced ~50% in GBMs of patientswith glomerulosclerosis (33), and similarchanges have been demonstrated in ex-tramural coronary vessels of diabetic pa-tients (34). In diabetic animals, the syn-thesis, concentration, and sulfation ofHS-PG are significantly decreased(35,36). But, is it possible that thesequalitative changes of the extracellularmatrix (i.e., decreased density of nega-

tively charged HS-PG) are the commoncause of microalbuminuria and the en-hancement of mesangial expansion andpremature atherosclerosis?

HS-PG is synthesized in the en-dothelial, mesangial, and myomedialcells. After sulfation has taken place inthe Golgi apparatus (a process imitatedby the enzyme N-deacetylase), HS-PG isincorporated into the extracellular ma-trix of glomeruli and large arteries, whereit contributes to the structural integrityof the basal membrane and vessel walls(37).

HS-PG ANDALBUMINURIA— Decreased densityof HS within the GBM leads to albumi-nuria (38,83). This also was seen in arecent study that demonstrated that in-jection of monoclonal antibodies againstHS induces albuminuria within minutes(83). Furthermore, when researchers in-creased the density of HS-PG in glomer-uli of diabetic rats by ACE inhibitors, theprogression of albuminuria was arrested(39). Thus, in theory, it is not unlikelythat the decreased density of anionicHS-PG demonstrated in glomeruli of di-abetic patients might be the cause of theincreased glomerular filtration of plasmaalbumin. HS-PG also contributes tokeeping the mesangial (40,41) and myo-medial (42) cells in a resting state. De-creased concentration of HS in tissue cul-tures of mesangial cells has been shownto lead to decreased inhibition and sub-sequently to proliferation of mesangialcells (40,41).

Table 4—Similarities of structural and functional alterations of glomeruli and largevessel walls in albuminuric patients with diabetes mellitus

ALTERATION REFERENCE

ACCUMULATION OF EXTRACELLULAR MATRIX

DECREASED DENSITY OF HS-PG

PROLIFERATION OF MESANGIAL AND MYOMEDIAL CELLS

INCREASED POSTENDOTHELIAL MACROMOLECULAR

PERMEABILITY ACROSS THE EXTRACELLULAR MATRIX

(34, 82)(33, 34)(40-42)

(49; BENT-HANSEN, UNPUBLISHED

OBSERVATIONS)


Dow



Table 5—Alteration in diabetic patients with early nephropathy, possibly related todecreased density of HS-PG within the extracellular matrix

NORMAL UAER INCREASED UAER

All differences are statistically highly significant.

REFERENCE

RENAL CLEARANCE RATIO OF

NEUTRAL VS. ANIONIC

MACROMOLECULES

TERA (°/o X H"1)P-TRIGLYCERIDES (MM)

P-VLDL (MM)P-FIBRINOGEN (^-M)

P-VON WILLEBRAND FACTOR

(IU/ML)

IN VIVO PLATELET ADHESION (%)

2.5

5.20.980.577.620.86

36

1.0

7.91.281.079.621.20

46

(27)

(27, 49)(26)(26)(26)(47)

(48)

4

-I- f- fDo

<3O

Dine

30-100

Figure 4—TERA (percentage per hour) incontrol subjects (N) and IDDM patients withdifferent levels of UAER. Do, long-term diabeticswith normal UAER; Dinc, incipient diabeticnephropathy; Dn, clinical diabetic nephropathy.From Deckert et al. (27). ® by Diabetologia.

HS-PG ANDATHEROSCLEROSIS— In large ves-sel walls, HS-PG has anti-atherogeneticproperties (43), and a negative correla-tion has been demonstrated between HSand the cholesterol content of humanlarge vessel walls (44). HS-PG contrib-utes to the structural integrity of the in-terendothelial clefts of large vessel walls.These interendothelial clefts contributeto the transendothelial pathway for lipo-proteins and other macromolecules.HS-PG also helps to maintain the myo-medial cells in a resting state and, inlarge vessel walls, the anionic HS-PGcontributes to binding of lipoprotein li-pase and antithrombin III (reviewed[43]). A decreased density of anionicHS-PG in endothelial plasma membranesand extracellular matrices, as suggestedin diabetic patients with microalbumin-uria, is expected to lead to a decreasedbinding of lipoprotein lipase (45) andtherefore to a decreased clearance of tri-glycerides and VLDLs. Increased plasmaconcentration of triglycerides and VLDLsis seen in IDDM patients with albumin-uria (26). Decreased binding of anti-thrombin III would shift the hemostaticbalance in a procoagulant direction,leading to deposition of fibrin on theendothelial cells' surface. Deposition offibrin, however, leads to the release ofvon Willebrand factor (46). Increased

concentration of von Willebrand factor isseen in diabetic patients with albumin-uria (47). Increased density of vonWillebrand factor on endothelial cell sur-face will increase the in vivo platelet ad-hesion that also is seen in diabetic pa-tients with albuminuria (48). Most signsand symptoms in patients with incipientnephropathy are compatible with de-creased density of anionic HS-PG inextracellular matrices (27) (Table 5).

DECREASED ANIONIC HS-PG:THE CAUSE OF ALBUMINURIAAND PREMATUREATHEROSCLEROSIS?— Is the de-creased density of HS-PG in the GBMand large vessel walls actually the causeof microalbuminuria and premature ath-erosclerosis in diabetes? Accumulatingevidence suggests that this might be thecase. Most of this evidence comes fromcareful studies of the very first clinicalsigns of diabetic nephropathy, namelymicroalbuminuria and its concomitantincreased TERA, a marker of postendo-thelial macromolecular permeability.The coincidence of microalbuminuriaand TERA is rather striking (27) (Fig. 4).The transendothelial flux of albumin inpatients with microalbuminuria is in-creased by —50%, compared with long-term diabetic patients with normalUAER and similar metabolic control

(49). Is the increased UAER and the in-creased TER of albumin caused by de-creased density of anionic HS-PG? In-creased TER can be attributable toincreased intravascular pressure, in-creased number and sizes of vascularmacromolecular pathways, or composi-tional changes of the macromolecularpathway. Increased intravascular pres-sure seems not to be the cause becauseblood pressure (27), capillary pressure(E. Tooke, unpublished observations),oncotic pressure differences (50), andplasma flow were not different betweenpatients with normal UAER and patientswith persistently increased UAER(51,52). Also an increased number andsize of macromolecular pathways haveuntil now not been demonstrated in pa-tients with microalbuminuria. Thus, vas-cular volume (50), capillary density (52),capillary surface (52), and even the num-ber and width of interendothelial clefts,are not different between patients withand without microalbuminuria (T.Deckert and Egeberg, unpublished ob-servations). These results might suggestthat increased TERA in patients with mi-croalbuminuria is caused by qualitativechanges of the macromolecular pathwayand the extracellular matrix. Structural-functional studies of the vascular extra-cellular matrix are difficult to perform invivo. However, the extracellular matrix


Dow


Microalbuminuria in micro- and macro/vascular disease

EPITHELIUM

LAMINA

RARA EXTERNA

LAMINA DENSA

LAMINARARA INTERNAENDOTHELIUM

oTYPE IV COLLAGENt LAMININo HEPARAN SULPHATE% PLASMA ALBUMIN

Figure 5—Structure of GBM. Q , type IV collagen; x, laminin; O, HS-PG (negatively charged); • , plasma albumin (negatively charged).

of the glomerular filtration barrier—theglomerular basement membrane(GBM)—can be studied more easily.

The composition of the GBM iswell-known (Fig. 5). The GBM is nega-tively charged, and HS-PG contributes tothis negative charge. The negative chargeinhibits the filtration of negativelycharged plasma proteins (i.e., exerts acharge selectivity). We have assessed re-nal charge selectivity in IDDM patientsby measuring the fractional clearance ofpairs of plasma proteins of identical sizebut different charges (27). We usednonglycosylated albumin and more neg-atively charged glycosylated albumin assuch a pair (53), and neutral total IgGand the more negative subfraction IgG4(27,54,55). With both pairs of plasmaproteins, we found loss-of-charge selec-tivity in patients with microalbuminuria.

This means that most nephrons of pa-tients with albuminuria are no longerable to differentiate between plasma pro-teins of different charges. Similar resultswere obtained by others using IgG andIgG4 (56) and pancreatic and salivaryisoamylase (57) for the assessment ofcharge selectivity. It has been suggestedthat loss-of-charge selectivity might beattributable to an increase of the largepore area of the GBM. However, studieswith dextran clearances indicate that thefractional dextran clearances in patientswith microalbuminuria are identical tothe fractional dextran clearance in non-diabetic and diabetic patients with nor-mal UAER (58; T. Deckert, et al., un-published observations). Thus, changesin size selectivity seem to be preceded bychanges in charge selectivity. Alterationin tubular function also is sometimes

suggested to bias the estimation of renalcharge selectivity. However, this is notvery likely because the findings of loss ofrenal charge selectivity would suggest in-creased negative charge of the tubularmembranes—even though no mecha-nism of such kind has been demon-strated. The excretion of P2"microgl°b-ulin and retinol binding protein—twomarkers of tubular function with slightlydifferent isoelectric points—are com-pletely unchanged in diabetic patientswith microalbuminuria compared withpatients with normal UAER (T. Deckertet al., unpublished observations). There-fore, we suggest that loss-of-charge se-lectivity in patients with microalbumin-uria is attributable to loss of negativelycharged components within the glomer-uli, probably HS-PG. A negative correla-tion is seen between the number of an-


Dow



Table 6—Anionic charge of HS-PG in extracellular matrix of diabetic patients is reducedby certain factors

DECREASED DENSITY OF HS-PG

INCREASED GENE EXPRESSION (60, 62) AND SECRETION (63) OF COLLAGEN IV AND FIBRONECTIN

DECREASED SYNTHESIS OF HS-PG (36)

DECREASED TURNOVER OF GLYCOSYLATED COLLAGEN IV (84)

DECREASED BINDING OF HS-PG TO GLYCOSYLATED COLLAGEN IV (69)

DECREASED SULFATION OF HS (65, 68)

ionic sites and albuminuria (59). Anegative correlation has been seen be-tween mRN A of HS-PG and albuminuriain diabetic mice (60). Furthermore, an-tibodies against HS will induce albumi-nuria (83), and ACE inhibitors given todiabetic rats has been shown to increasethe content of HS within the glomeruli,and simultaneously arrest albuminuria(39). From these studies, we concludedthat decreased density of anionic HS-PGmight be the cause of microalbuminuriain diabetic patients. Because the anioniccharge on the luminal surface of vascularwalls has been shown to be decreased indiabetics patients (61) and because wehave shown that charge selectivity ofTERA also was found to disappear inpatients with microalbuminuria (L. Bent-Hansen, et al., unpublished observa-tions), we suggest that—in the extracel-lular matrix of vascular walls—adecrease of anionic HS-PG is occurringalso, leading to an enhancement of theatherosclerotic processes.

CAUSES OF DECREASED HS-PGIN DIABETES:HYPERGLYCEMIA— How does de-creased density of anionic HS-PG occurin diabetic patients? Hyperglycemia leadsto decreased density of anionic HS-PG byseveral mechanisms (Table 6). The mostimportant cause seems to be a defect inthe coordinated genetic expression ofcomponents of the extracellular matrix.Cagliero et al. (62) have studied the bio-synthesis of extracellular matrix compo-nents in tissue cultures of human endo-thelial cells. They found that, in mostpatients, hyperglycemia leads to in-

creased expression of collagen IV, lami-nin, and fibronectin. The activation ofthese extracellular matrix componentsseem to be coordinated. When collagenIV is activated by 300% the transcriptionof fibronectin also is increased by~300%. Thus, hyperglycemia leads to acoordinated activation of the expressionand secretion (63) of collagen IV, lami-nin, and fibronectin. This does not seemto be the case for HS-PG. Therefore, theratio of mRNA for HS-PG/collagen IV issignificantly reduced in diabetic KK mice(60). Increased gene expression and se-cretion of collagen IV without alterationof the expression of HS-PG would ex-plain not only the thickening of base-ment membrane in diabetic patients, butalso a relative loss of anionic HS-PG inthe extracellular matrix.

Other mechanisms also affect thenegative charge of the extracellular ma-trix in diabetic patients. Thus, it recentlyhas been demonstrated that diabetes af-fects the activity of N-deacetylase, thekey enzyme in the sulfation of HS-PG(Fig. 6) (64-66). In poorly regulatedSTZ-diabetic Sprague-Dawley rats, theactivity of N-deacetylase in the liver andglomeruli is decreased by —50%, a re-duction that is correlated to mean bloodglucose. The activity can be normalizedby insulin treatment. In Sprague-Dawleyrats, the reduction in N-deacetylase ac-tivity is correlated to albuminuria (66).Decreased activity of N-deacetylase leadsto undersulfation of HS, which meansloss of negative charges. Undersulfationof HS in diabetic animals has been dem-onstrated by several groups (35,68).

A third mechanism by which di-

abetes might affect the density of HS-PGin the extracellular matrix is nonenzy-matic glycosylation. In poorly regulateddiabetic patients, the degree of glycosyl-ation of collagen IV and fibronectin ismuch higher than in control subjects(69). However, nonenzymatic glycosyl-ation of collagen IV and fibronectin willdecrease the affinity to HS-PG (69). Thismight lead to a greater turnover ofHS-PG in poorly regulated diabetic pa-tients (70) and reduced density ofHS-PG in extracellular matrix. Becausepatients with albuminuria often havepoor diabetes control and higher HbAlc

than long-term diabetic patients withnormal UAER, it is likely that thesemechanisms also contribute to decreaseddensity of HS-PG. Thus, diabetes affectsthe biosynthesis of the extracellular ma-trix in several ways and leads to reduceddensity of anionic HS-PG in the extracel-lular matrix. Besides poor diabetes con-trol, genetic factors seem to influence themetabolism of extracellular matrix com-ponents.

GENETIC FACTORS— Epidemiologi-cal studies have demonstrated that only aminority of diabetic patients will developclinical nephropathy and its associatedcardiovascular complications (71). Thus,the majority of diabetic patients seem tobe protected from the development ofdiabetic nephropathy. This protectioncannot be explained exclusively by bettermetabolic control (72). Probably severalsusceptibility genes are involved. Recentstudies have demonstrated a clustering ofdiabetic nephropathy within families thathave several cases of diabetes amongfirst-degree relatives (73,74). Other re-cent studies suggested a genetic defect inthe coordinated secretion of glycosami-noglycans among patients with albumin-uria. These studies demonstrated that fi-broblasts from diabetic patients withnephropathy kept in tissue cultures forseveral months under identical conditionsecreted relatively less HS comparedwith hyaluronic acid than fibroblasts


Dow



EXTRACELLULAR MATRIXSUSCEPTIBILITY AND HYPERGLYCAEMIA

BLOODNORMO-GLYCAEMIA

NON-SUSCEPTIBLE SUSCEPTIBLE

HYPERGLYCAEMIANON-SUSCEPTIBLE SUSCEPTIBLE

MESANGIAL/MYOMEDIAL CELLS

Figure 6—Extracellular matrix in glomenili and large vessel walls in normoglycemic (left) and hyperglycemic (right) individuals. N-deacetylase is thekey enzyme in the sulfation of HS-PG (note branching symbols). For explanation, see text.

from control subjects, whereas fibro-blasts from diabetic patients with normalUAER were similar to control subjects(75). Furthermore, experiments in dia-betic rats of slightly different geneticbackground indicated differences in theregulation of the N-deacetylase as a sus-ceptibility factor (65,66). Thus, besidespoor diabetes control, genetic differencesin the regulation of the biosynthesis ofthe extracellular matrix might contributeto individual differences in the composi-tional structure of the matrix and therebyto susceptibility to microalbuminuriaand its associated complications. The hy-pothesis that emerges from these obser-vations is as follows.

In nondiabetic patients not sus-ceptible for albuminuria and prematureatherosclerosis, collagen IV and HS-PGsynthesis are normally regulated, result-ing in a normal composition of the extra-cellular matrix. In susceptible normogly-cemic individuals, however, defects inthe regulation of HS-PG biosynthesis willresult in a reduced density of HS-PG,

turning these individuals into subjectsmore vulnerable to albuminuria and ath-erosclerosis. In diabetic patients, thedensity of HS-PG is reduced mostly be-cause of a relative increased synthesis ofcollagen IV. HS-PG is also undersulfatedbecause of reduced N-deacetylase activ-ity. The net alterations are, therefore, de-pendent mainly on the quality of meta-bolic control. In susceptible diabeticpatients, however (i.e., patients with ge-netic defects in the regulation of HS-PG),extracellular matrix synthesis will resultin a more pronounced reduction of thedensity and sulfation of HS-PG withinplasma membranes and the extracellularmatrix. In these patients, the biochemicalalterations may lead to functional alter-ations, such as increased permeability andproliferation of mesangial and myomedicalcells, and finally to the well-known struc-tural alterations of glomerulosclerosis andpremature atherosclerosis.

Because DNA probes of HS-PGand N- deacetylase are or will be availablein the future, it is hoped that some of the

susceptibility genes will be identifiedsoon. Then, identification of patients atrisk might be possible. By concentratingour clinical efforts on these patients, it ishoped that prevention of some of themost severe complications of diabetesmight, in the future, be more successful.

References1. Borch-Johnsen K, Kragh Andersen P,

Deckert T: The effect of proteinuria onrelative mortality in type 1 (insulin-dependent) diabetes mellitus. Diabetolo-gia 28:590-96, 1985

2. Parving H-H, Smidt UM, Andersen AR,Svendsen PA: Early aggressive antihyper-tensive treatment reduces rate of declinein kidney function in diabetic nephrop-athy. Lancet 1:1175-79, 1983

3. Borch-Johnsen K, Kreiner S: Proteinuria:value as predictor of cardiovascular mor-tality in insulin dependent diabetes mel-litus. BrMedJ 294:1651-54, 1987

4. Jensen T, Borch-Johnsen K, Kofoed-Enevoldsen A, Deckert T: Coronary heartdisease in young type 1 (insulin-depen-


Dow


Dechert and Associates

dent) diabetic patients with and withoutdiabetic nephropathy: incidence and riskfactors. Diabetologia 30:144-48, 1987

5. Marshall SM: Screening for microalbumi-nuria: which measurement? Diabetic Med8:706-11, 1991

6. Lind B, Jensen T, Feldt-Rasmussen B,Deckert T: Normal urinary albumin ex-cretion in recently diagnosed type 1 di-abetic patients. Diabetic Med 6:682-84,1989

7. Christensen CK, Mogensen CE: Thecourse of incipient diabetic nephropathy:studies of albumin excretion and bloodpressure. Diabetic Med 2:97-102, 1985

8. Mathiesen ER, R0nn B, Jensen T, StormB, Deckert T: Relationship betweenblood pressure and urinary albumin ex-cretion in development of microalbumi-nuria. Diabetes 39:245-49, 1990

9. Mogensen CE, Chacati A, ChristensenCK, Close CF, Deckert T, Hommel E,Kastrup J, Lefebvre P, Mathiesen ER,Feldt-Rasmussen B, Schmitz A, VibertiGC: Microalbuminuria: an early markerof renal involvement in diabetes. UremiaInvest 9:85-95, 1986

10. Niazy S, Feldt-Rasmussen B, Deckert T:Microalbuminuria in insulin-dependentdiabetes: prevalence and practical conse-quences. JDiabet Complications 1:76-80,1987

11. Gall M-A, Rossing P, Skott P, Damsbo P,Vaag A, Bech K, Dejgaard A, LauritzenM, Lauritzen E, Hougaard P, Beck-Nielsen H, Parving HH: Prevalence ofmicro- and macroalbuminuria, arterialhypertension, retinopathy and large ves-sel disease in European type 2 (non-insulin-dependent) diabetic patients. Di-abetologia 34:655-61, 1991

12. Walker JD, Viberti GC: Aetiology andpathogenesis of diabetic nephropathy:clues from early functional abnormali-ties. In Textbook of Diabetes. Pickup J,Williams G, Eds. Oxford, UK, BlackwellScientific Publications, 1991, p. 657-71

13. Schmitz A, Vaeth M: Microalbuminuria:a major risk factor in non-insulin-dependent diabetes: a 10-year follow-upstudy of 503 patients. Diabetic Med5:126-34, 1988

14. YudkinJS, Forrest RD, Jackson CA: Mi-croalbuminuria as predictor of vascular

disease in non-diabetic subjects. Lancet1:530-33, 1988

15. Damsgaard EM, Fr0land A, j0orgensenOD, Mogensen CE: Microalbuminuria aspredictor of increased mortality in el-derly people. Br Med J 300:297-300,1990

16. Earle KA, Walker JD, Jones SL, Hill C,Viberti GC: Cardiovascular disease in theparents of type 1 (insulin-dependent) di-abetics with nephropathy (Abstract). Di-abetologia 33:A66, 1990

17. N0rgaard K, Mathiesen ER, Hommel E,Jensen JS, Parving H-H: Lack of familialpredisposition to cardiovascular diseasein type 1 (insulin-dependent) diabeticpatients with nephropathy. Diabetologia34:370-72, 1991

18. Krolewski AS, Canessa M, Warram JH,Laffel LMB, Christlieb AR, Knowler WC,Rand LI: Predisposition of hypertensionand susceptibility to renal disease in in-sulin-dependent diabetes mellitus. NEnglJ Med 318:140-45, 1988

19. Viberti GC, Keen H, Wiseman MJ: Raisedarterial pressure in parents of proteinuricinsulin-dependent diabetics. Br Med J295:515-17, 1987

20. Jensen JS, Mathiesen ER, N0rgaard K,Hommel E, Borch-Johnsen K, Funder J,Brahm J, Parving H-H, Deckert T: In-creased blood pressure and erythrocytesodium/lithium countertransport activityare not inherited in diabetic nephrop-athy. Diabetologia 33:619-24, 1990

21. Walker JD, Tariq T, Viberti GC: Sodium-lithium countertransport activity in redcells of patients with insulin dependentdiabetes and nephropathy and their par-ents. Br Med J 301:635-38, 1990

22. Mangili R, Bending JJ, Scott G, Li KL,Gupta A, Viberti GC: Increased sodium-lithium countertransport activity in redcells of patients with insulin-dependentdiabetes and nephropathy. N Engl J Med318:1146-50, 1988

23. Elving LD, Wetzels JFM, de Nobel E,Berder JHM: Erythrocyte sodium-lithiumcountertransport is not different in type 1(insulin-dependent) diabetic patientswith and without diabetic nephropathy.Diabetologia 34:126-28, 1991

24. Davies JE, Ng LL, Li LK, Earl KL, Kofoed-Enevoldsen A, Trevisan R, Viberti GC:

Intracellular pH and Na"VH+ antiportactivity of cultured fibroblasts in diabeticnephropathy. Diabetic Med (BDA Ab-stracts):39A, 1991

25. Chase HP, Garg SK, Marshall G, Berg C ,Harris S, Jackson WE, Hamman RE: Cig-arette smoking increases the risk of albu-minuria among subjects with type 1 di-abetes. JAMA 265:614-17, 1991

26. Jensen T, Stender S, Deckert T: Abnor-malities in plasma concentrations of lip-oproteins and fibrinogen in type 1 (insu-lin-dependent) diabetic patients withincreased urinary albumin excretion. Di-abetologia 31:142-45, 1988

27. Deckert T, Feldt-Rasmussen B, Borch-Johnsen K, Jensen T, Kofoed-EnevoldsenA: Albuminuria reflects widespread vas-cular damage: the Steno hypothesis. Di-abetologia 32:219-26, 1989

28. Nelson RG, Kunzelman CL, Pettitt DJ,Saad MF, Bennett PH, Knowler WC: Al-buminuria in type 2 (non-insulin-depen-dent diabetes mellitus) and impaired glu-cose tolerance in Pima Indians.Diabetologia 32:870-76, 1989

29. Mattock MB, Keen H, Viberti GC, El-Gohari MR, Murrells TJ, Scott GS, WingJR, Jackson PG: Coronary heart diseaseand urinary albumin excretion rate intype 2 (non-insulin-dependent) diabeticpatients. Diabetologia 31:82-87, 1988

30. Niskanen L, Uusitupa M, Sarlund H, Si-itonen O, Voutilainen E, Penttila I, Py-6'rala K: Microalbuminuria predicts thedevelopment of serum lipoprotein ab-normalities favoring atherogenesis innewly diagnosed type 2 (non-insulin-dependent) diabetic patients. Diabetolo-gia 33:237-43, 1990

31. Schmitz A, IngerslevJ: Haemostatic mea-sures in type 2 diabetic patients withmicroalbuminuria. Diabetic Med 7:521-25, 1990

32. 0sterby R, Parving H-H, Hommel E, J0r-gensen HE, L0kkegaard H: Glomerularstructure and function in diabeticnephropathy: early to advanced stagesDiabetes 39:1057-63, 1990

33. Shimomura H, Spiro RG: Studies onmacromolecular components of humanglomerular basement membrane and al-terations in diabetes: decreased levels ofheparan sulfate proteoglycan and lami-


Dow



nin. Diabetes 36:374-81, 198734. Dybdahl H, Ledet T: Diabetic macroan-

giopathy. Quantitative histopathologicalstudies of the extramural coronary arter-ies from type 2 (non-insulin-dependent)diabetic patients. Diabetologia 30:882-86, 1987

35. Kjellen L, Bielefeld D, Hook M: Reducedsulfation of liver heparan sulfate in ex-perimentally diabetic rats. Diabetes 32:337-42, 1983

36. Wu V-Y, Wilson B, Cohen MP: Distur-bances in glomerular basement mem-brane glycosaminoglycans in experimen-tal diabetes. Diabetes 36:679-83, 1987

37. Gallagher JT, Lyon M, Steward WP:Structure and function of heparan sul-phate proteoglycans. BiochemJ 236:313-25, 1986

38. Rosenzweig LJ, Kan war YS: Removal ofsulfated (heparan sulfate) or nonsulfated(hyaluronic acid) glucosaminoglycans re-sults in increased permeability of the glo-merular basement membrane to 1 2 5 1 -bovine serum albumin. Lab Invest 47:177-83, 1982

39. Reddi A, Ramamurthi R, Miller M, Dhu-per S, Lasker N: Enalapril improves al-buminuria by preventing glomerular lossof heparan sulfate in diabetic rats. Bio-chemical Med Metab Biol 45:119-31,1991

40. Castellot JJ, Hoover RL, Harper PA, Kar-novsky MJ: Heparin and glomerular epi-thelial cell-secreted heparin-like speciesinhibit mesangial-cell proliferation. Am]Pathol 120:427-35, 1985

41. Kitamura M, Mitarai T, Maruyama N,Nagasawa R, Yoshida H, Sakai O: Mesan-gial cell behavior in a three-dimensionalextracellular matrix. Kidney Int 40:653-61, 1991

42. Pomerantz KB, Hajja DP: Eicosanoids inregulation of arterial smooth muscle cellphenotype, proliferative capacity andcholesterol metabolism. Arteriosclerosis9:413-29, 1989

43. Deckert T, Jensen T, Feldt-Rasmussen B,Kofoed-Enevoldsen A, Borch-Johnsen K,Stender S: Albuminuria, a risk marker ofatherosclerosis in insulin dependent dia-betes mellitus. Cardiovasc Risk Factors

1:347-60, 199144. Hollmann J, Schmidt A, von Bassewitx DB,

Buddecke E: Relationship of sulfated gly-cosaminoglycans and cholesterol contentin normal and arteriosclerotic humanaorta. Arteriosclerosis 9:154-58, 1989

45. Braun JEA, Severson DL: Diabetes re-duces heparin- and phospholipase C-re-leasable lipoprotein lipase from cardio-myocytes. Am J Physiol 260:E447-85,1991

46. Ribes JA, Ni F, Wagner DD, Francis CW:Mediation of fibrin-reduced release ofvon Willebrand factor from cultured en-dothelial cells by the fibrin (3 chain. J ClinInvest 84:43335-342, 1989

47. Jensen T: Increased plasma concentra-tion of von Willebrand factor in insulindependent diabetics with incipientnephropathy. BrMedJ 298:27-28, 1989

48. Valdorf-Hansen F, Jensen T, Borch-Johnsen K, Deckert T: Cardiovascularrisk factors in type 1 (insulin-dependent)diabetic patients with and without pro-teinuria. Ada Med Scand 222:439-44,1987

49. Feldt-Rasmussen B: Increased transcapil-lary escape rate of albumin in type 1(insulin-dependent) diabetic patientswith microalbuminuria. Diabetologia 29:282-86, 1986

50. Hommel E, Mathiesen ER, Aukland K,Parving H-H: Pathophysiological aspectsof edema formation in diabetic nephrop-athy. Kidney Int 38:1187-92, 1990

51. Feldt-Rasmussen B, Baker L, Deckert T:Exercise as a provocative test in earlyrenal disease in type 1 (insulin-depen-dent) diabetes: albuminuric, systemicand renal haemodynamic responses. Di-abetologia 28:389-96, 1985

52. Leinonen H, Matikainen E, JuntunenJ:Permeability and morphology of skeletalmuscle capillaries in type 1 (insulin-dependent) diabetes mellitus. Diabetolo-gia 22:158-62, 1982

53. Kverneland A, Feldt-Rasmussen B, VidalP, Welinder B, Bent-Hansen L, S0egaardU, Deckert T: Evidence of changes inrenal charge selectivity in patients withtype 1 (insulin-dependent) diabetes mel-litus. Diabetologia 29:634-39, 1986

54. Deckert T, Feldt-Rasmussen B, Djurup R,Deckert M: Glomerular size and chargeselectivity in type 1 (insulin-dependent)diabetes mellitus. Kidney Int 33:100-

106, 198855. Deckert T, Feldt-Rasmussen B, Djurup R,

Deckert M: Glomerular size and chargeselectivity in type 1 (insulin-dependent)diabetes mellitus (Abstract). Diabetologia30:513A, 1987

56. Pietravalle P, Morano S, Christina G, DeRossi MG, Mariani G, Cotroneo P,Ghirlanda G, Clementi A, Andreani D, DiMario U: Charge selectivity of protein-uria in type 1 diabetes explored by im-munoglobulin subclass clearance (Ab-stract). Diabetes 40 (Suppl. 1):552A,1991

57. FoxJG, QuinJD, Paterson KR, O'ReillyDS, Boulton-Jones JM: Impaired glomer-ular charge selectivity in microalbumin-uria and diabetic nephropathy (Ab-stract). Diabetologia 33 (Suppl.):A67,1990

58. Nakamura Y, Myers BD: Charge selectiv-ity of proteinuria in diabetic glomerulop-athy. Diabetes 37:1202-11, 1988

59. Vernier RL, Steffes MW, Sisson-Ross S,Mauer SM: Heparan sulfate proteoglycanin the glomerular basement membrane intype 1 diabetes mellitus. Kidney Int 41:1070-80, 1992

60. Ledbetter S, Copeland EJ, Noonan D,Vogeli G, Hassell JR: Altered steady-statemRNA levels of basement membraneproteins in diabetic mouse kidneys andthromboxane synthase inhibition. Diabe-tes 39:196-203, 1990

61. Raz 1, Havivi Y, Yarom R: Reduced neg-ative surface on arterial endothelium ofdiabetic rats. Diabetologia 31:618-20,1988

62. Cagliero E, Roth T, Roy S, Lorenzi M:Characteristics and mechanisms of high-glucose-induced overexpression of base-ment membrane components in culturedhuman endothelial cells. Diabetes 40:102-110, 1991

63. Haneda M, Kikkawa R, Horide N, To-gawa M, Koya D, Kajiwara N, OoshimaA, Shigeta Y: Glucose enhances type IVcollagen production in cultured rat glo-merular mesangial cells. Diabetologia 34:198-200, 1991

64. Unger E, Pettersson I, Eriksson UJ, Lin-dahl U, Kjellen L: Decreased activity ofthe heparan sulfate modifying enzymeglucosaminyl N-deacetylase in hepato-


Dow



cytes from streptozotocin-diabetic rats. JBiol Chem 266:8671-74, 1991

65. Kofoed-Enevoldsen A, Eriksson UJ: Inhi-bition of N-acetyl-heparosan deacetylasein diabetic rats. Diabetes 40:1449-52,1991

66. Kofoed-Enevoldsen A: Inhibition of glo-merular glucosaminyl N-deacetylase indiabetic rats. Kidney Int. 41:763-67,1992

68. Cohen MP, Klepser H, Wu V-Y: Under-sulfation of glomerular basement mem-brane heparan sulfate in experimental di-abetes and lack of correction with aldosereductase inhibition. Diabetes 37:1324-27, 1988

69. Tarsio JF, Reger LA, Furcht LT: Molecu-lar mechanisms in basement membranecomplications of diabetes: alterations inheparin, laminin, and type IV collagenassociation Diabetes 37:532-39, 1988

70. Klein DJ, Oegema TR, Jr., Brown DM:Release of glomerular heparan 35SO4

proteoglycan by heparin from glomeruliof streptozocin-induced diabetic rats. Di-abetes 38:130-39, 1989

71. Kofoed-Enevoldsen A, Borch-Johnsen K,Kreiner S, Nerup J, Deckert T: Decliningincidence of persistent proteinuria intype 1 (insulin-dependent) diabetic pa-tients in Denmark. Diabetes 36:205-209, 1987

72. Deckert T, PoulsenJE: Diabetic nephrop-

athy: fault or destiny? Diabetologia 21:178-83, 1981

73. Seaquist ER, Goetz FC, Rich S, BarbosaJ:Familial clustering of diabetic kidneydisease. N Engl J Med 320:1161-65,1989

74. Borch-Johnsen K, Norgaard K, HommelE, Mathiesen ER, Jensen JS, Deckert T,Parving H-H: Diabetic nephropathy—aninherited complication? Kidney Int. 41:719-22, 1992

75. Deckert T, Horowitz IM, Kofoed-Enevoldsen A, Kjellen L, Deckert M, Lyk-kelund C, Burcharth F: Possible geneticdefects in regulation of glycosaminogly-cans in patients with diabetic nephrop-athy. Diabetes 40:764-70, 1991

76. Kofoed-Enevoldsen A, Jensen T, Borch-Johnsen K, Deckert T: Incidence of reti-nopathy in type 1 (insulin-dependent)diabetes: association with clinicalnephropathy. J Diabet Complications3:96-99, 1987

77. Kelbaek H, Jensen T, Feldt-Rasmussen B,Christensen NJ, Richter EA, Deckert T:Impaired left-ventricular function in in-sulin-dependent diabetic patients withincreased urinary albumin excretion.ScandJ Clin Invest 51:467-73, 1991

78. Jensen T, Richter EA, Feldt-RasmussenB, Kelbaek H, Deckert T: Impaired aero-bic work capacity in insulin dependentdiabetics with increased urinary albumin

excretion. BrMedJ 196:1352-54, 198879. Bent-Hansen L, Deckert T: Metabolism

of albumin and fibrinogen in type 1 (in-sulin-dependent) diabetes mellitus. Dia-betes Res 7:159-64, 1988

80. Jensen T, Bjerre-Knudsen J, Feldt-Rasmussen B, Deckert T: Features of en-dothelial dysfunction in early diabeticnephropathy. Lancet 1:461-63, 1989

81. Feldt-Rasmussen B, Mathiesen ER, Deck-ert T, Giese J, Christensen NJ, Bent-Hansen L, Nielsen MD: Central role forsodium in the pathogenesis of bloodpressure changes independent of angio-tensin, aldosterone and catecholaminesin type 1 (insulin-dependent) diabetesmellitus. Diabetologia 30:610-17, 1987

82. Ellis EN, Steffes MW, Goetz FC, Suther-land DER, Mauer SM: Glomerular filtra-tion surface in type 1 diabetes mellitus.Kidney M 29:889-94, 1986

83. Van Den Born J, Van den Heuvel LPWJ,Bakker MAH, Veerkamp JH, AssmannKJM, Berden JHM: A monoclonal anti-body against GBM heparan sulfate in-duces an acute selective proteinuria inrats. Kidney lnt 41:115-23, 1992

84. Brownlee M, Cerami A, Vlassara H: Ad-vanced glycosylation end products in tis-sue and the biochemical basis of diabeticcomplications. N Engl J Med 318:1315-21, 1988


Dow


microalbuminuria - diabetes care

Documents