Nephrol Dial Transplant (1998) 13: 1056–1079 NephrologyDialysis

TransplantationAbstracts

Proteinuria and progressive renal disease: towards preventing renal diseaseFourth International Symposium, 24–25 September 1997, Montreux,SwitzerlandOrganizers: Michel Burnier, Switzerland; Geoffrey Boner, Israel; Hans Brunner, Switzerland;David van Dijk, Israel; Ab Donker, The Netherlands; Carl Erik Mogensen, Denmark

trials in non-diabetic patients, suggests that dietary proteinIntroductionrestriction indeed delays the onset of end-stage renal disease[1]. Evidence also suggests that dietary protein restriction

A series of meetings on Proteinuria and Progressive retards the progression of diabetic renal disease as well [2].Renal Disease has been conducted over the past few The Modification of Diet in Renal Disease Study, a largeyears in Paris (1992), Vienna (1994), Amsterdam multicenter clinical trial also demonstrates a benefit of dietary(1996) and the fourth meeting in Montreux 1997. protein restriction and aggressive reduction in blood pressure

on the rate of progression of chronic renal disease in humansProceedings of the third meeting in Amsterdam werewith significant proteinuria and baseline ongoing deteriora-published in Nephrology Dialysis Transplantation 1997;tion of renal function [3]. Angiotensin converting enzyme12 [Suppl 2]: 1–85.(ACE) inhibitors have also been shown to offer protection inThe proceedings of the fourth meeting are organizedpatients with diabetic and non-diabetic chronic renal diseaseas extended abstracts, and provide a overview of the[4,5].different approaches towards progressive renal disease, Among diabetics, randomized multicenter clinical trials

both in diabetic and non-diabetic nephropathy. have examined the merits of ACE inhibitors in both type 1Special attention is given to the concept of micro- and type 2 patients. Evidence in support of a strong renopro-

albuminuria, with special reference to early treatment, tective effect of this class of antihypertensive therapy waswithin the area of diabetes. obtained in those with already overt nephropathy as well as

in those with incipient nephropathy [4,6 ]. Based on theseThe organizers warmly thank the sponsors of thefindings various professional organizations have issuededucational grants: Astra Hassle, Bayer, Boehringerformal guidelines enthusiastically supporting ACE inhibitorsMannheim, Bristol–Myers Squibb, Hoechst Marionin diabetic subjects. Studies are also in progress to determineRoussel, Janssen–Cilag, Knoll, Merck Sharp &whether angiotensin II receptor antagonists are equivalent orDohme, Novo, Servier, and Zeneca.possibly more renoprotective than ACE inhibitors. Theresults of these trials will not be available, however, for atCarl Erik Mogensenleast 3 years.

1. Fouque D, Laville M, Boissel J, Chiffet R, La Beeuw M, ZechPY. Controlled low protein diets in chronic renal insufficiency:Progression of renal diseasemeta-analysis. Br Med J 1992; 304: 216–220

Barry Brenner; Renal Division, Brigham and Women’s 2. Zeller K, Whittaker E, Sullivan L, Raskin P, Jacobson HR. EffectHospital and Harvard Medical School, Boston, MA of restricting dietary protein on the progression of renal failure

in patients with insulin-dependent diabetes mellitus. N Engl J MedGlomerular hypertension has been implicated as a major1991; 324: 78–84factor in mediating progressive renal damage after any of a 3. Levey A, Adler S, Caggiula A, England B et al. Effects of dietary

variety of initiating injuries. Amelioration of glomerular protein restriction on the progression of advanced renal diseasehypertension by dietary protein restriction or antihypertens- in the Modification of Diet in Renal Disease (MDRD) Study.ive therapy reduces progressive glomerular damage in experi- Am J Kidney Dis 1996; 27: 652–663

4. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect ofmental models of chronic renal disease. Glomerularangiotensin-converting-enzyme inhibition on diabetic nephro-hypertension and hyperfiltration also occur in humans withpathy. N Engl J Med 1993; 329: 1456–1462diabetes mellitus, solitary or remnant kidneys, and various

5. Maschio G, Alberti D, Janin G, Locatelli F, Mann JF et al.forms of acquired renal disease. Therapies shown to limitEffect of the angiotensin-converting-enzyme inhibitor benazeprilglomerular hypertension and to ameliorate renal injury inon the progresson of chronic renal insufficiency. N Engl J Medanimals have begun to be tested in humans. Dietary protein 1996; 334: 939–945

restriction and antihypertensive drug therapy have been the 6. Viberti G, Mogensen CE, Groop LC, Pauls JF. Effect of captoprilprincipal therapeutic interventions tested thus far. A meta- on progression to clinical proteinuria in patients with insulin-analysis of studies of low-protein diets in renal insufficiency, dependent diabetes mellitus and microalbuminuria. JAMA 1994;

271: 275–279based only on data from prospective randomized, controlled

© 1998 European Renal Association–European Dialysis and Transplant Association

Abstracts 1057

Parving H-H, Hommel E, Mathiesen E et al. Prevalence of micro-Population-based studies of microalbuminuria and renalalbuminuria, arterial hypertension, retinopathy and neuropathydiseasein patients with insulin dependent diabetes. Br Med J 1988;

Bo Feldt-Rasmussen; Medical Department P 2132, 296: 156–160Rigshospitalet, Copenhagen Ø, Denmark Deckert T, Yokoyama H, Mathiesen E et al. Cohort study of

predictive value of urinary albumin excretion for atheroscleroticMicroalbuminuria is present when the urinary albumin excre-vascular disease in patients with insulin dependent diabetes. Brtion rate (UAER) in a 24-h urine or a short-time collected Med J 1996; 312: 871–874

urine during day time is in the range of 30–300 mg/24-h(20–200 mg/min equivalent to 0.46–4.6 mmol/24 h). A UAERin the microalbuminuric range is a strong predictor ofnephropathy in insulin-dependent diabetes (type 1). Becauseof this close association between microalbuminuria and pro- Proteinuria and glomerular heparan sulphate alterationsgressive renal disease in type 1 diabetic patients this associ-

C. J. Ilse Raats, Jacob Van Den Born, Jo H. M. Berden;ation has been intensely studied. Also the epidemiology ofDivision of Nephrology, University Hospital St Radboud,microalbuminuria and renal disease has been investigated.Nijmegen, The NetherlandsFrom a number of epidemiological studies it appears that the

prevalence of microalbuminuria is 12–22%. There are contro-versies as to the relation between prevalence of microalbumi- Introductionnuria and diabetes duration. A number of studies observe an Heparan sulphate (HS) is the anionic glycosaminoglycanalmost unchanged prevalence of microalbuminuria independ- side-chain of heparan sulphate proteoglycan (HSPG) presentent of diabetes duration. Most of these studies have included in basement membranes, in extracellular matrix and on cellonly one urine sample for classification of the albuminuria surfaces. The impact of HS for the permselective propertieslevel. This may constitute a problem because of the high day of the glomerular basement membrane (GBM) was docu-to day variation of UAER. Studies including more than one mented by a number of observations. First, enzymatic diges-urine sample describe a three fold increase in prevalence of tion of HS by heparitinase caused an enhanced permeabilitymicroalbuminuria after 24 years of diabetes compared to 5–9 of the GBM for native ferritin and albumin [1]. Second, anyears of diabetes. acute, selective proteinuria could be induced in rats by a single

Despite the close relationship between microalbuminuria intravenous injection of a monoclonal antibody directedand diabetic nephropathy it has been shown that the kidney against GBM HS [2]. Third, a reduction in GBMfunction in terms of glomerular filtration rate declines only in HS-associated anionic sites was demonstrated with cationicthe patients with a UAER high in the microalbuminuric probes in several human and experimental proteinuric glom-range. Furthermore, microalbuminuria has also in type 1 dia- erulopathies [3]. With recently developed antibodies directedbetic patients shown to be an independent risk factor not only against the GBM HSPG core protein and the HS side chainof renal disease but also of cardiovascular disease. The latter [4], we demonstrated a decrease in HS staining in the GBMassociation seems to be even more prominent in patients with in different human proteinuric glomerulopathies, such as sys-non-insulin dependent diabetes (type 2) as well as in non temic lupus erythematosus (SLE), minimal change disease,diabetic apparently healthy controls. The prevalence of mic- membranous glomerulonephritis and diabetic nephropathy,roalbuminuria is as high as 30–40% in type 2 diabetic patients whereas the staining of the HSPG core protein remainedand end-stage renal failure only occurs in 3–8% of these unaltered [5]. This suggested changes in the HS side-chains ofpatients. The strong link between microalbuminuria and dia- HSPG. To get more insight into the mechanisms responsiblebetic nephropathy in type 1, therefore, may also be mediated for this observation, we studied GBM HS expression in vari-through mechanisms similar to those associating micro- ous experimental models of proteinuria. From these investi-albuminuria to cardiovascular disease in type 1 and type 2 gations four different mechanisms have emerged until nowdiabetic patients as well as in apparently healthy subjects. which are discussed below, namely masking of HS by immune

complexes, depolymerization of HS by radicals, proteolyticcleavage of HSPG by enzymes released from inflammatoryOrchard TJ, Dorman JS, Maser RE et al. Prevalence of complica-cells and metabolically induced biochemical changes of HS.tions in IDDM by sex and duration. Diabetes 1990; 39: 1116–1124

Gall M-A, Rossing P, Skøtt P et al. Prevalence of micro- andmacroalbuminuria, arterial hypertension, retinopathy and large

Masking of HS by immune complexesvessel disease in European type 2 (non-insulin-dependent) diabeticpatients. Diabetologia 1991; 34: 655–661 SLE is an autoimmune disease characterized by the occur-

Klein R, Klein BEK, Linton KLP, Moss SE. Microalbuminuria in rence of numerous autoantibodies, primarily directed againsta population-based study of diabetes. Arch Intern Med 1992; nuclear antigens. In both human [5] and murine [6 ] SLE152: 153–158 nephritis, albuminuria is paralleled by a decrease in GBM HSFeldt-Rasmussen B, Borch-Johnsen K, Deckert T, Jensen G, Jensen

staining and an increase in immune complex depositions inJS. Microalbuminuria: An important diagnostic tool. Journal ofthe glomerular capillary wall. This decrease in HS stainingDiabetes and Its Complications 1994; 8: 137–145was inversely correlated with both albuminuria and withThe EURODIAB IDDM Complications Study Group.

Microvascular and acute complications in IDDM patients. immune complex deposition. Formerly, it was thought thatDiabetologia 1994; 37: 278–285 anti-DNA autoantibodies could cross-react with HS in the

Joner G, Brinchmann-Hansen O, Torres C G, Hanssen KF. A GBM. Later we could show, however, that antinuclear anti-nationwide cross-sectional study of retinopathy and microalbumi- bodies are able to bind to HS and to the GBM via nucleo-nuria in young Norwegian type 1 (insulin-dependent) diabetic somal antigens [7]. The positively charged histones in thesepatients. Diabetologia 1992; 35: 1049–1054 nucleosome-containing immune complexes interact with theDamsgaard EM, Mogensen CE. Microalbuminuria in elderly hyper-

negatively charged HS in the GBM. Despite the decrease inglycaemig patients and controls. Diabetic Med 1986; 3: 430–435HS staining, the HS content in glomeruli of lupus mice isMogensen CE, Damsgaard EM, Frøland A et al. Microalbuminurianormal, which suggests that HS is masked by immune com-in non-insulin-dependent diabetes. Clin Nephrol 1992: 38: [Suppl.

1]: 528–538 plexes. This is supported by the finding that the binding of a

Abstracts1058

monoclonal anti-HS antibody to HS in ELISA can be inhib- heparanase, released from leukocytes and phagocytes uponstimulation, can degrade HS in extracellular matrix [15].ited by autoantibodies complexed to nucleosomes but not by

purified non-complexed autoantibodies [6 ]. Therefore, in Therefore, in forms of glomerulonephritis that are accompan-ied by inflammatory cell influx, albuminuria may be the resultSLE nephritis and possibly also in other glomerular diseases

in which immune complexes are known to play a role in the of digestion of HS by released lysosomal enzymes.pathogenesis, deposition of immune complexes in the glomer-ular capillary may lead to masking of HS, resulting in an

Biochemical changes of HSenhanced permeability of the GBM for macromolecules.Indeed, inhibition of binding of nucleosome-complexed In human diabetic nephropathy, albuminuria is accompaniedautoantibodies to the GBM by heparin (oids) prevents both by a decrease in GBM HS staining [16 ]. In streptozotocinthe decrease of HS staining in the GBM and albuminuria in (STZ)-induced diabetes in rats, the number of HS-associatedMRL/lpr mice [8]. anionic sites in the lamina rara externa of the GBM is reduced

whereas there is no difference in absolute GBM HS contentbetween control and diabetic rats. This may point to struc-

Depolymerization of HS by radicals tural alterations in HS, most likely undersulphation [16,17].Adriamycin (ADR) nephropathy, an experimental model for HS undersulphation has also been shown for other tissues ofthe nephrotic syndrome, is characterized by heavy albumin- STZ-induced diabetic rats. Culture of mesangial cells on mes-uria and hypalbuminemia. Also in this model, there is a angial matrix that was non-enzymatically glycated or pro-decrease in GBM HS staining which is inversely correlated longed exposure of mesangial cells to elevated glucose levelswith albuminuria without changes in the staining of the leads to a decreased production of HS which is under-HSPG core protein and HS attachment sites. Treatment of sulphated [18]. The undersulphation of HS could be due to arats with ADR nephropathy with the hydroxyl radical scav- decreased activity of the enzyme glucosaminyl N-deacetylase,enger dimethylthiourea (DMTU ) results in amelioration of the key enzyme in HS sulphation. Indeed, in hepatocytes ofalbuminuria and partial prevention of HS loss from the STZ-induced diabetic rats, a reduced activity of this enzymeGBM, indicating a role for hydroxyl radicals in this model. was found [19]. Undersulphation leads to a disturbed inter-An explanation for the decrease of HS staining in ADR action of HS with other GBM components and might lead tonephropathy is the in vitro observation that HS is depolymer- an increased permeability of the GBM.ized by hydroxyl radicals. This depolymerization of HS byreactive oxygen species (ROS) in vitro can be prevented byDMTU. Taken together, these data suggest that in ADR Conclusionnephropathy HS is depolymerized by hydroxyl radicals lead- From the above-mentioned observations in various experi-ing to loss of GBM integrity and albuminuria [9]. mental glomerulopathies it can be concluded that different

Polymorphonuclear leukocyte (PMN )-derived ROS may mechanisms can lead to an impairment of HS function andhave similar effects. Activated PMNs that are attached to thus of the charge-dependent permeability of the GBM. Sinceextracellular matrix secrete ROS and lysosomal enzymes. In HS interacts with other constituents of the GBM, such asthe early phase of passive anti-GBM nephritis, PMN-derived collagen type IV, laminin and fibronectin, any change in HSROS and lysosomal enzymes are responsible for the glomer- can also lead to a disturbed integrity of the GBM leading toular damage. In this model, treatment with scavengers of an increased size-dependent permeability. Besides the effectsROS was shown to ameliorate albuminuria, possibly via pre- on GBM permeability, changes in HS may have consequencesvention of HS depolymerization [10,11]. Also nitric oxide for other processes involved in glomerular pathology, like(NO) can degrade HS and heparin in vitro via the formation growth factor activity (bFGF/PDGF), mesangial expansionof HNO2 and the cleavage of glycosaminoglycans by cultured and proliferation and local coagulation. Therefore, these HSendothelial cells can be prevented by inhibition of NO pro- changes may aggrevate glomerular inflammation.duction [12]. Because NO is produced by endothelial cells,

Acknowledgements. Part of this research was supported byPMNs and monocytes after inflammatory stimulation, it maythe Dutch Kidney Foundation (C93.1318 and C95.1513).be involved in HS depolymerization and the developmentDr J. van den Born is supported by the Dutch Diabetesof albuminuria in immunologically mediated glomerularFund. The scientific contributions of Mrs Marinka Bakkerinflammation.and Dr Mieke van Bruggen are gratefully acknowledged.

Cleavage of HS (PG) by enzymes 1. Rosenzweig LJ, Kanwar YS. Removal of sulfated (heparansulfate) or nonsulfated (hyaluronic acid) glycosaminoglycansBesides the injurious effects of PMN-derived ROS, PMNsresults in increased permeability of the glomerular basementalso release proteases upon activation. Amoung the enzymesmembrane to 125I-bovine serum albumin. Lab Invest 1982;released are the cationic serine proteases elastase and cathep- 47: 177–184

sin G which are known to digest fibronectin, laminin and 2. Van den Born J, Van den Heuvel LPWJ, Bakker MAH,collagen type IV. The involvement of these proteases in the Veerkamp JH, Assmann KJM, Berden JHM. A monoclonaldevelopment of albuminuria is substantiated by the observa- antibody against GBM heparan sulfate induces an acute selective

proteinuria in rats. Kidney Int 1992; 41: 115–123tion that beige mice, whose PMNs are deficient for elastase3. Cotran RS, Rennke HG. Anionic sites and the mechanisms ofand cathepsin G, do not develop albuminuria although they

proteinuria [editorial ]. N Engl J Med 1983; 309: 1050–1052show a comparable influx of PMNs as control mice [13].4. Van den Born J, Van den Heuvel LPWJ, Bakker MAH,Furthermore, in vivo perfusion of elastase in rats results in

Veerkamp JH, Assmann KJM, Berden JHM. Monoclonal anti-proteinuria and decrease in GBM HS staining and to a lesserbodies against the protein core and glycosaminoglycan sideextent of HSPG core protein staining. In vitro experiments chain of glomerular basement membrane heparan sulfate proteo-

showed that elastase, which is highly cationic at physiological glycan: characterization and immunohistological application inpH, can bind to the anionic HS. Next, elastase can cleave human tissues. J Histochem Cytochem 1994; 42: 89–102HSPG near the HS attachment sites and HS side-chains 5. Van den Born J, Van den Heuvel LPWJ, Bakker MAH et al.

Distribution of GBM heparan sulfate proteoglycan core proteinbound to small peptide fragments are released [14]. Also

Abstracts 1059

and side chains in human glomerular diseases. Kidney Int 1993; rises within the normal range. This observation raises the43: 454–463 possibility that elevated pressure levels may be one important

6. Van Bruggen MCJ, Kramers K, Hylkema MN et al. Decrease factor contributing to the development of renal damage.of heparan sulfate in the glomerular basement membrane in Indeed, diabetic nephropathy clusters in families and diabeticmurine lupus nephritis. Am J Pathol 1995; 146: 753–763 patients who develop albuminuria belong to families with a

7. Kramers C, Hylkema MN, Van Bruggen MCJ et al. Anti- higher prevalence of arterial hypertension and cardiovascularnucleosome antibodies complxed to nucleosomal antigens showdisease. Of interest diabetic patients with microalbuminuriaanti-DNA reactivity and bind to rat glomerular basement mem-are more insulin-resistant than patients with normal albuminbrane in vivo. J Clin Invest 1994; 94: 568–577excretion rate and this may explain why patients with mic-8. Van Bruggen MCJ, Walgreen B, Rijke GPM et al. Heparin androalbuminuria tend to have a poorer metabolic control andheparinoids prevent the binding of immune complexes containing

nucleosomal antigens to the GBM and delay nephritis in higher arterial pressures. Insulin resistance is an independentMRL/lpr mice. Kidney Int 1996; 50: 1555–1564 risk factor for coronary artery disease in the general

9. Raats CJI, Bakker MAH, Van den Born J, Berden JHM. population.Hydroxyl radicals depolymerize glomerular heparan sulfate in Thus diabetic patients with microalbuminuria have avitro and in experimental ne phrotic syndrome. J Biol Chem cluster of risk factors for both renal and cardiovascular1997; in press complications and follow-up studies have shown that10. Rehan A, Johnson KJ, Wiggins RC, Kunkel RG, Ward PA.

microalbuminuria is one of the strongest predictors forEvidence for the role of oxygen radicals in acute nephrotoxiccardio-renal disease both in type 1 and type 2 diabetes.nephritis. Lab Invest 1984; 51: 396–403

11. Feith GW, Assmann KJM, Bogman MJJT, Van Gompel APM,Yip J, Mattock M, Sethi M, Morocutti A, Viberti GC. InsulinSchalkwijk J, Koene RAP. Different mediator systems in

resistance in insulin-dependent diabetic patients with microalbum-biphasic heterologous phase of anti-GBM nephritis in mice.inuria. Lancet 1993; 342: 883–887Nephrol Dial Transplant 1996; 11: 599–607

Microalbuminuria Collaborative Study Group. Risk factors for the12. Vilar RE, Ghael D, Li M et al. Nitric oxide degradation ofdevelopment of microalbuminuria in insulin-dependent diabeticheparin and heparan sulphate. Biochem J 1997; 324: 473–479patients: a cohort study. Br Med J 1993; 306: 1235–123913. Schrijver G, Schalkwijk J, Robben JCM, Assmann KJM, Koene Nelson RG, Pettitt DJ, Baird HR et al. Prediabetic blood pressureRAP. Antiglomerular basement membrane nephritis in beige predicts urinary albumin excretion after the onset of type 2 (non-

mice. J Exp Med 1989; 169: 1435–1448 insulin-dependent) diabetes mellitus in Pima Indians. Diabetologia14. Heeringa P, Van den Born J, Brouwer E et al. Elastase, but not 1993; 36: 998–1001

proteinase 3 (PR3), induces proteinuria associated with loss of Haneda M, Kikkawa R, Togawa M, Koya D, Kajiwara N, Uzu T.glomerular basement membrane heparan sulphate after in vivo High blood pressure is a risk factor for the development ofrenal perfusion in rats. Clin Exp Immunol 1996; 105: 321–329 microalbuminuria in Japanese subjects with non-insulin-

15. Vlodavsky I, Eldor A, Haimovitz-Friedman A et al. Expression dependent diabetes mellitus. J Diabetic Complications 1992; 6:of heparanase by platelets and circulating cells of the immune 181–185system: possible involvement in diapedesis and extravasation. Messent J, Elliott T, Hill R, Jarrett J, Keen H, Viberti GC.Invas Metast 1992; 12: 112–127 Prognostic significance of microalbuminuria in insulin-dependent

16. Tamsma JT, Van den Born J, Bruijn JA et al. Expression of diabetes mellitus: A twenty-three year follow-up study. Kidney Intglomerular extracellular matrix components in human diabetic 1992; 41: 836–839nephropathy: decrease of heparan sulphate in the glomerular Mattock MB, Morrish NJ, Viberti GC, Keen H, Fitzgerald AP,basement membrane. Diabetologia 1994; 37: 313–320 Jackson G. Prospective study of microalbuminuria as predictor

17. Van den Born J, Van Kraats AA, Bakker MAH et al. Reduction of mortality in NIDDM. Diabetes 1992; 41: 736–741of heparan sulphate-associated anionic sites in the glomerular Barzilay J, Warram JH, Bak M, Laffel LMB, Canessa M, Krolewskibasement of rats with streptozotocin-induced diabetic nephro- AS. Predisposition to hypertension: risk factor for nephropathypathy. Diabetologia 1995; 38: 1169–1175 and hypertension in IDDM. Kidney Int 1992; 41: 723–730

18. Van Det NF, Van den Born J, Tamsma JT et al. Effects of high Quinn M, Angelico MC, Warram JH, Krolewski AS. Familialglucose on the production of heparan sulfate proteoglycan by factors determine the development of diabetic nephropathy inmesangial and epithelial cells. Kidney Int 1996; 49: 1079–1089 patients with IDDM. Diabetologia 1996; 39: 940–945

19. Unger E, Petterson I, Eriksson UJ, Lindahl U, Kjellen L. Viberti, GC, Keen H, Wiseman MJ. Raised arterial pressure inDecreased activity of the heparan sulfate-modifying enzyme parents of proteinuric insulin-dependent diabetics. Br Med J 1987;glucosaminyl N-deacetylase in hepatocytes from streptozotocin- 295: 5157diabetic rats. J Biol Chem 1991; 266: 8671–8674 Krolewski AS et al. Predisposition to hypertension and susceptibility

to renal disease in insulin-dependent diabetes mellitus. N EnglJ Med 1988; 318: 140–145

Earle K, Walker J, Hill C, Viberti GC. Familial clustering ofcardiovascular disease in patients with insulin-dependent diabetesand nephropathy. N Engl J Med 1992; 326: 673–677Detecting the Risk of Renal and Cardiovascular Disease in

DiabetesGianCarlo Viberti; Division of Medicine, Guy’s andSt Thomas’s Medical and Dental School, London,UK

Antihypertensive treatment in microalbuminuric type 2Elevated albumin excretion rate is the earliest sign of diabeticdiabetic patientsnephropathy. There is a consistent and independent associ-Juliana C. N. Chan, Julian A. J. H. Critchley, Clive S.ation of microalbuminuria with higher levels of blood pres-Cockram; Department of Medicine and Therapeutics, Thesure, though often within the normal range, which exceeds byChinese University of Hong Kong, The Prince of Walesapproximately 10–15 percent the blood pressure levels ofHospital, Shatin, Hong Konglong-term diabetic patients with normoalbuminuria. This

phenomenon has been elegantly documented in studies which Both hypertension and type 2 diabetes mellitus are heterogen-have used 24hr blood pressure monitoring. Studies of trans- eous diseases. Blood pressure is determined by cardiac outputition from normo- to microalbuminuria have demonstrated and systemic vascular resistance, both of which are influencedthat those diabetic patients who progress already show by multiple interacting hormonal, vasoactive and growth fac-

tors. Hypertension in diabetic patients, including those withincreases in blood pressure while the albumin excretion rate

Abstracts1060

type 2 disease, have evidence of salt retention as shown [15] and 2 diabetic patients required more than one anti-hypertensive drug for optimal control of blood pressure [23].by increased plasma atrial natriuretic peptide and reduced

plasma aldosterone and renin concentrations [1]. These Given the complex nature of diabetes, these findings raise theimportant question as to which combination therapy will pro-patients also have enhanced vascular responsiveness to exo-

genous administration of catecholamines and angiotensin II vide the optimal antihypertensive, renal and metabolic effects.Although an ACE inhibitor plus a low dose diuretic is a(AII ) [2]. Elevated plasma concentrations of these hormones

and serum angiotensin converting enzyme (ACE) activity, popular choice of therapy, the combination of an ACEinhibitor and a calcium channel blocking agent has potentialsuggesting activation of these hormonal systems, have also

have been reported [1]. On the other hand, reduced natriur- additional advantages. Calcium channel blocking agents arepotent antihypertensive agents given either alone or in com-etic responses, such as reduced urinary dopamine excretion,

may also contribute to this salt-retaining tendency [3]. In bination with an ACE inhibitor [24,25] especially in patientswith salt retention [26 ], such as diabetic [27] and elderlyaccord with this pathophysiological heterogeneity, different

classes of antihypertensive drugs with different modes of patients [28]. In some patient groups, such as those withyoung age or suboptimal metabolic control, a calcium chan-actions have been shown to be effective in reducing blood

pressure in diabetic patients, albeit also with marked hetero- nel blocking agent may be a preferred agent given the poten-tial adverse effects of a diuretic such as those on intermediarygeneity in treatment responses [4]. The latter is further

influenced by potential inter-ethnic pharmacokinetic and metabolism and sexual dysfunction, albeit small and dose-related [13]. Despite the extensive clinical use of calcium-pharmacodynamic differences [5,6 ].

Hypertension and proteinuria frequently coexist in type 2 channel blocking agents, the renal effects of this class ofagents remain controversial [29]. Although the dihydropyrid-diabetic patients, especially in non-Caucasian populations

[7]. The coexistence of these two conditions markedly ine calcium-channel blocking agents, such as nifedipine andamlodipine, may increase proteinuria due to their dilatingincrease the risk of cardiovascular death and renal failure in

these patients. The pathogenesis of diabetic proteinuria is effects on both afferent and efferent glomerular arterioles,their beneficial effects on GFR due to a reduction of systemiccomplex and involves vascular (systemic and intraglomerular

hypertension), metabolic (hyperglycaemia and hyperlipidae- blood pressure should be taken into consideration [29].Despite the potential cardioprotective effects of blockingmia), growth and genetic factors [8]. However, control of

glycaemia [9] and blood pressure [10] are two of the most agents [30] and the beneficial effects of blocking agents oninsulin resistance [12], there are no long-term studies examin-powerful determinants for the progression of proteinuria. In

patients with clinical proteinuria and established renal dis- ing the use of these agents in type 2 diabetic patients. On theother hand, the new class of antihypertensive agent, the AIIease, adequate control of blood pressure is of paramount

importance [10]. As shown in one of the meta-analyses, a type 1 receptor antagonist, such as losartan, has been shownto reduce proteinuria more effectively than a calcium-channelreduction of mean arterial pressure of 10 mmHg is accompan-

ied by an increase of 3.7 ml/min/year of glomerular filtration blocking agent despite similar, and in some patients, lessreduction in blood pressure [31]. There is now an ongoingrate (GFR) [4]. However, since proteinuria and GFR are also

determined by metabolic factors such as hyperglycaemia and 4-year study, the RENAAL (Reduction in Endpoints inNIDDM with AII Antagonist, Losartan) Study, which exam-hyperlipidaemia [11], the metabolic effects of antihypertens-

ive agents may influence their renal effects. In this respect, in ines the renoprotective effects of losartan versus placebo intype 2 diabetic patients with macroalbuminuria and renalview of their relatively neutral effects on carbohydrate and

lipid metabolism, ACE inhibitors and calcium channel impairment. However, more studies are needed to define theoptimal antihypertensive regimen in type 2 diabetic patientsblocking agents are frequently used as first line antihyper-

tensive agents in diabetic patients [12–14]. who make up the majority of patients with renal failure. Inthis respect, a randomized, parallel study examining the com-There is now a wealth of data confirming the anti-

proteinuric effects of ACE inhibitors in both type 1 and type bination of an ACE inhibitor with a low-dose diuretic or acalcium channel blocking or an AII receptor antagonist on2 diabetic patients, an effect which is independent of blood

pressure reduction [4]. The use of this class of drug has also vascular, renal and metabolic indices as well as clinical end-points will be of particular interest and major therapeuticbeen shown to reduce the rate of decline in renal function in

type 1 [15], normotensive type 2 diabetic [16,17] as well as relevance.non-diabetic patients with chronic renal failure [18,19].Although there are sufficient reasons to believe that these 1. Chan JCN, Cheung CK, Cockram CS, Critchley JAJH,

Swaminathan R, Nicholls MG. Atrial natriuretic peptide (ANP)beneficial renal effects of ACE inhibitors should also beand renin-angiotensin-aldosterone system in patients with non-applicable to hypertensive type 2 diabetic patients, definitiveinsulin-dependent diabetes (NIDDM). J Hum Hypertens 1994;proof in this important therapeutic area is still lacking. To8: 451–456date, there have only been a few long term studies, lasting

2. Weidmann P, Ferrari P. Central role of sodium in hypertensionmore than one year, which examined the renal effects ofin diabetic subjects. Diabetes Care 1991; 14: 220–232

different antihypertensive agents in type 2 diabetic patients 3. Chan JCN, Critchley JAJH, Ho CS, Nicholls MG, Cockramwith hypertension [20–23]. In all of these studies comparing CS, Swaminathan R. Atrial natriuretic peptide (ANP) andan ACE inhibitor with a calcium-channel blocking agent, no urinary dopamine output in non-insulin-dependent diabetesdifferences have been found between these 2 classes of drugs. mellitus. Clin Sci 1992; 83: 247–253

4. Kasiske BL, Kalil RS, Ma JZ, Liao M, Keane WF. Effect ofIn a 3-year study, hypertensive type 2 diabetic patients exhib-antihypertensive therapy on the kidney in patients with diabetes:ited a biphasic pattern in the deterioration in renal function.a meta-regression analysis. Ann Intern Med 1993; 118: 129–138In the first 12 months, the rate of decline in GFR correlated

5. Zhou HH, Koshakji RP, Silbertstein DJ, Wilkinson GR, Woodwith the reduction in blood pressure. This was followed by aAJJ. Racial differences in drug response. Altered sensitivity tostabilisation of GFR which correlated inversely with the and clearance of propranolol in men of Chinese descent as

reduction in blood pressure [23]. These findings further compared to American whites. N Engl J Med 1989; 320: 565–570emphasise the importance of conducting long term studies in 6. Tomlinson B, Anderson PJ, Chan JCN, Chan TYK, Critchleythe evaluation of renal function. JAJH. Ethnic differences in response to ACE inhibitors. In:

MacGregor GA, Sever PS, eds. Asian-Pacific Symposium onMost studies have shown that the majority of both type 1

Abstracts 1061

ACE Inhibition. Hong Kong Convention and Exhibition R, Cockram CS. Factors determining the blood pressureresponse to enalapril and nifedipine in hypertension associatedCentre, 1993with NIDDM. Diabetes Care 1995; 18: 1001–10067. Tierney MW, McDonald CJ, Luft FC. Renal disease in hyper-

28. Buhler FR, Hulthen UL, Kiowski W, Boli P. Greater antihyper-tensive adults: effect of race and type II diabetes mellitus. Amtensive efficacy of the calcium channel inhibitor verapamil inJ Kidney Dis 1989; 13: 485–493older and low renin patients. Clin Sci 1982; 63: S439-S4428. Parving H-H, Tarnow L, Rossing P. The angiotensin-converting

29. Zanchi A, Brunner HR, Waeber B, Burnier M. Renal haemo-enzyme gene and its inhibition in diabetic nephropathy. Currdynamic and protective effects of calcium antagonists in hyper-Opin Endocrinol Diabetes 1996; 3: 315–321tension. J Hypertens 1995; 13: 1363–13759. The Diabetes Control and Complications Trial Research Group.

30. The MIAMI trial research group. Metoprolol in acute myocard-The effect of intensive treatment of diabetes on the developmential infarction (MIAMI). A randomized placebo-controlled inter-and progression of long-term complications in insulin-dependentnational trial. Eur Heart J 1985; 6: 199–211diabetes mellitus. N Engl J Med 1993; 329: 977–986

31. Chan JCN, Critchley JAJH, Tomlinson B, Chan TYK, Cockram10. Parving HH, Andersen AR, Smidt UM, Hommel E, MathiesenCS. Antihypertensive and anti-albuminuric effects of losartanER, Svendsen PA. Effect of antihypertensive treatment on kidneypotassium and felodipine-ER in Chinese elderly hypertensivefunction in diabetic nephropathy. Br Med J 1987; 294: 1443–1447patients with or without NIDDM. Am J Nephrol 1997; 17: 72–8011. Ravid M, Neumann L, Lishner N. Plasma lipids and the

progression of nephropathy in diabetes mellitus type II: effectof ACE inhibitors. Am J Kidney Dis 1995; 47: 907–910

12. Lithell HOL. Effect of antihypertensive drugs on insulin, glucose,and lipid metabolism. Diabetes Care 1991; 40: 203–209

Glycaemic control, islet function and the development and13. Chan JCN, Cockram CS, Critchley JAJH. Drug-induced dis-orders of glucose metabolism—mechanisms and management. progression of diabetic nephropathyDrug Safety 1996; 15: 135–156 Michael Steffes; Department of Laboratory Medicine and

14. Chan JCN, Yeung VTF, Leung DHY, Tomlinson B, Nicholls Pathology, University of Minnesota, USAMG, Cockram CS. The effects of enalapril and nifedipine oncarbohydrate and lipid metabolism in NIDDM. Diabetes Care A multicentre study of the efficacy of intensive therapy on the1994; 17: 859–862 complications of diabetes mellitus, the Diabetes Control and

15. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect of Complications Trial (DCCT) enrolled 1441 patients, nearlyangiotensin converting enzyme inhibition on diabetic nephro- all of whom had normal renal function. After an averagepathy. N Engl J Med 1993; 329: 1456–1462 follow-up of 5–6 years the DCCT Research Group concluded16. Ravid M, Savin H, Jutrin I, Bental T, Katz B, Lishner M.

in a decisive manner that intensive diabetic therapy reducedLong-term stabilizing effect of angiotensin-converting enzymethe risk of all diabetic complications (by about 50% for eachinhibition on plasma creatinine and on proteinuria in normotens-complication) [1,2]. This conclusion, while of great positiveive type II diabetic patients. Ann Intern Med 1993; 118: 577–581import to the diabetic patient, places the emphasis upon17. Ravid M, Lang R, Bachmani R, Lishner M. Long term renopro-therapeutic application to achieve optimal glycaemic control.tective effect of angiotensin converting enzyme inhibition in non-

insulin dependent diabetes mellitus. A 7 year follow up study. In nearly all studies prospectively evaluating the develop-Arch Intern Med 1996; 156: 286–289 ment and progression of diabetic nephropathy, the diabetic

18. The GISEN Group. Randomised placebo-controlled trial of patients who progress to nephropathy have higher haemo-effect of ramipril on decline in glomerular filtration rate and globin A1c values than those diabetic patients who did notrisk of terminal renal failure in proteinuric, non-diabetic nephro- advance [3,4]. Thus it is difficult to demonstrate the independ-pathy. Lancet 1997; 349: 1857–1863 ence of other factors thought to predict or indicate the

19. Maschio G, Alberti D, Janin G, Locatelli F, Mann J, Moroleselikelihood or the actual presence of diabetic nephropathy.M, et al. Effect of the angiotensin converting enzyme inhibitor

The familial influence in the development of diabeticbenazepril on the progression of chronic renal insufficiency. Nnephropathy was greatly strengthened by the DCCTEngl J Med 1996; 334: 939–945Research Group, who demonstrated a relationship between20. Ruggenenti P, Mosconi L, Bianchi L, Cortesi L, Campana M,DCCT probands and relatives for the development of retino-Pagani G, et al. Long-term treatment with either enalapril or

nitrendipine stabilizes albuminuria and increases glomerular pathy and nephropathy [5]. This relationship was most easilyfiltration rate in non-insulin-dependent diabetic patients. Am shown in the standard treatment group, with similar trendsJ Kidney Dis 1994; 24: 753–761 but a diminished influence in families of patients receiving

21. Chan JCN, Cockram CS, Nicholls MG, Cheung CK, intensive therapy. Thus these observations re-emphasizeSwaminathan R. Comparison of enalapril and nifedipine in glycaemic control as fundamental cause of microvasculartreating non-insulin dependent diabetes associated with hyper- complications.tension: one year analysis. Br Med J 1992; 305: 981–985 Several investigators over the past 2 decades have shown22. Melbourne Diabetic Nephropathy Study Group. Comparison

microalbuminuria to be the primary indicator/predictor ofbetween perindopril and nifedipine in hypertensive and normot-diabetic nephropathy [6 ]. Its relationship to the rise in bloodensive diabetic patients with microalbuminuria. Br Med J 1991;pressure has not always been clear; however, in the DCCT302: 210–216microalbuminuria (verified in at least 2 measurements) occurs23. Velussi M, Brocco E, Frigato F, Zolli M, Muollo B, Maioli M,2 years prior to the rise in blood pressure. DCCT data alsoet al. Effects of cilazapril and amlodipine on kidney function in

hypertensive NIDDM patients. Diabetes 1996; 45: 216–222 demonstrated that increasing the level to designate estab-24. Morgan T, Anderson A, Bopper J. Enalapril and nifedipine in lished microalbuminuria enhances its specificity as an indic-

essential hypertension; synergism of the hypotensive effects in ator or predictor of diabetic nephropathy (e.g. albuminuriacombination. Clin Exp Hypertens (A) 1988; 10: 779–789 of 30 mg/24 h is somewhat less specific than 50 mg/24 h).

25. Guazzi MD. Calcium channel blockade with nifedipine and The presence of clinical albuminuria (>300 mg/24 h) indi-angiotensin converting enzyme inhibition with captopril in the cates a clear risk for renal failure. The rate at which thetherapy of patients with severe primary hypertension. Circulation

glomerular filtration rate falls in patients with clinical diabetic1984; 70: 279–284nephropathy has been reduced with careful implementation26. Zanchetti A, Leonetti G. Discussion on the natriuretic effect ofof antihypertensive medication. Furthermore observationscalcium antagonists. J Cardiovasc Pharmacol 1987; 10: [Suppl.from pancreas transplantation indicate optimal glycemic con-1]: S161–S164

27. Chan JCN, Nicholls MG, Cheung CK, Law LK, Swaminathan trol may require several years before efficacy may be demon-

Abstracts1062

production, glycemic control and prevalence of microvascularstrated [7,8]. However, with careful implementation oflesions and polyneuropathy in long-term type 1 (insulin-depend-therapeutic modalities to improve glycemic control and toent) diabetes mellitus. Diabetologia 1987; 30: 208control hypertension, kidneys of diabetic patients may experi-

ence many more years of function before failing.In the practice of diabetes care basal insulin secretion or its

replacement with long-acting insulin enables efforts toapproach near-normal glycaemia. Yet in most type 1 diabetic Preservation of normal GFR in type 1 diabetic patients withpatients the clear benefit of a sustained level of basal insulin microalbuminuria under long-term (8 years) ACE inhibitionsecretion (usually indicated by a rise in c-peptide levels follow-

Elisabeth R. Mathiesen, Eva Hommel, Henrik P. Hansen,ing mixed meal stimulation) cannot be duplicated by exogen-Hans-Henrik Parving; Steno Diabetes Center, Gentofte,ous insulin therapy [9]. The DCCT Research GroupDenmarkdemonstrated that intensive glycaemic control extends the

time of demonstrable islet function [10]. The aim was to asses the long term effectiveness of angio-The DCCT Research Group also reported a lower risk for tensin converting enzyme inhibition on preservation of glom-

of retinopathy and nephropathy for those patients with resid- erular filtration rate (GFR) in type 1 diabetic patients withual islet function who received intensive therapy compared to microalbuminuria. Forty-four normotensive insulin depend-intensively treated patients with no residual islet function. Of ent diabetic patients with persistent microalbuminuria (urin-great interest was the accompanying reduced risk of hypogly- ary albumin excretion (UAE) 30–300 mg/24 h) were enrolledcaemia in those DCCT subjects with sustained c-peptide in an open randomized controlled study of 8 years duration.secretion receiving intensive treatment. Thus uniquely in The treatment group (n=21) was given captopril (50 mgthose patients with residual islet function the reduced risk of b.i.d.) and low-doze thiazide. The remaining 23 patients werethe microvascular complications was accompanied by a lower left untreated unless hypertension did develop (n=4). Allrisk of hypoglycaemia. except 2 patients from each group completed the 8-year

In general those patients who may more successfully reach follow-up. Sixteen of the 19 patients in the treatment groupthe goals of intensive therapy have an older age of onset and and 2 patients from the control group were subsequentlyare more likely to retain higher levels of c-peptide secretion investigated before and 2 months after cessation of antihyper-and thus sustained islet function. tensive treatment. The percentage of patients progressing to

nephropathy was 40% (9/23) in the control group and 10%(2/21) in the intervention group (P<0.02). During the pause1. DCCT Research Group. The effect of intensive treatment ofin captopril treatment a significant increase in UAE was dem-diabetes on the development and progression of long-term

complications in insulin-dependent diabetes mellitus. N Engl onstrated (P<0.001) and 6 of the 16 (38%) patients demon-J Med 1993; 329: 977–986 strated UAE above 300 mg/24 h. During the treatment pause

2. DCCT Research Group. Effect of intensive therapy on the GFR increased 8.6 (SE 3) ml/min (P<0.01) in the captoprildevelopment and progression of diabetic nephropathy in the group. From baseline to 8 years the fall in GFR was 12.2 (SEDiabetes Control and Complications Trial. Kidney Int 1995; 47: 6) ml/min (P<0.05) in the control group and 1,4 (2) (NS) in1703–1720 the captopril group (values measured during the treatment3. Parving HH, Hommel E, Mathiesen ER, Skøtt P, Edsberg B,

pause if possible) (P=0.06, between the groups). The fall inBahnsen M, Lauritzen M, Hougaard P, Lauritzen E. PrevalenceGFR during the study period in the 8 patients who developedof microalbuminuria, arterial hypertension, retinopathy andnephropathy was 25.6 (SE 10) (P<0.01) in the control groupneuropathy in patients with insulin dependent diabetes. Br Med

J 1988; 296: 156–160 and −3.5 (3) ml/min (NS) in the 6 patients from the capto-4. Microalbuminuria Collaborative Study Group. Risk factors for pril group with UAE >300 mg/24 h during the treatment

the development of microalbuminuria in insulin-dependent dia- pause (P<0.01 between the groups).betic patients: a cohort study. Br Med J 1993; 306: 1235–1239 Conclusion. The beneficial effect of angiotensin converting

5. Genuth S, Lachin J, Cleary P, Spielman R and the DCCT enzyme inhibition on postponing the development of diabeticResearch Group. Familial clustering of diabetes complications nephropathy is long lasting and associated with preservationin the DCCT. Abstract. Diabetes 1996; 46: suppl. 2: 30A

of normal GFR.6. Mogensen CE, Keane WF, Bennet PH, Jerums G, Parving HH,Steffes MW, Striker GE, Viberti GC. Screening and managementof microalbuminuria in patients with diabetes mellitus: recom- Mathiesen ER, Rønn B, Storm B, Foght H, Deckert T. The natural

course of microalbuminuria in insulin dependent diabetes: Amendations to the Scientific Advisory Board of the NationalKidney Foundation from an ad hoc committee of the Council 10-year prospective study. Diabetic Med 1995; 12: 482–487

Mathiesen ER, Feldt-Rasmussen B, Hommel E, Deckert T, Parvingon Diabetes Mellitus of the National Kidney Foundation. Lancet1995; 346: 1080–1085 H-H. Stable glomerular filtration rate in normotensive IDDM

patients with stable microalbuminuria—a 5 year prospective7. Bilous RW, Mauer SM, Sutherland DER, Najarian JS, GoetzFC, Steffes MW. The effects of pancreas transplantation on the study. Diabetes Care 1997; 20: 286–289

Rossing P, Hommel E, Smidt UM, Parving HH. Reduction inglomerular structure of renal allografts in patients with insulin-dependent diabetes. N Engl J Med 1989; 321: 80–85 albuminuria predicts diminished progression in diabetic nephro-

pathy. Kidney Int 1994; 45: 145–1498. Fioretto P, Mauer SM, Bilous RW, Goetz FC, Sutherland DER,Steffes MW. Effects of pancreas transplantation on glomerular Mathiesen ER, Hommel E, Giese J, Parving H-H. Efficacy of

captopril in postponing nephropathy in normotensive insulinstructure in insulin-dependent diabetic patients with their ownkidneys. Lancet 1993; 42: 1193–1196 dependent diabetic patients with microalbuminuria. Br Med J

1991; 303: 210–2169. Madsbad S, Krarup T, Regeur L, Faber, OK, Binder C. Effectof strict blood glucose control on residual beta cell function in Viberti G, Mogensen CE, Groop LC, Pauls JF. Effect of captopril

on progression to clinical proteinuria in patients with insulininsulin-dependent diabetics. Diabetologia 1981; 20: 530–53410. Steffes M, Tamborlane W, Becker D, Palmer J, Cleary P for the dependent diabetes mellitus and microalbuminuria. JAMA 1994;

271: 275–279DCCT Research Group. The effect of intensive diabetes treat-ment on residual beta cell function in the Diabetes Control and Laffel LMB, McGill JB, Gans DJ. The beneficial effect of

angiotensin-converting enzyme inhibition with captopril onComplications Trial (DCCT). Abstract. Diabetes 1996; 46:[Suppl. 2]: 18A diabetic nephropathy in normotensive IDDM patients with micro-

albuminuria. Am J Med 1995; 99: 497–50411. Sjoberg S, Gunnarsson R, Gjotterberg M et al. Residual insulin

Abstracts 1063

The Microalbuminuria Captopril Study Group. Captopril reduces Analysis of the influence of ACE inhibition on the well knownthe risk of nephropathy in IDDM patients with microalbuminuria. risk factors for the progression of renal disease demonstratesDiabetologia 1996; 39: 587–593 a very modest reduction of mean blood pressure, below stat-

istical significance; a modest decline in plasma total and LDLcholesterol and no change in HDL and HbAlc values. Themost important effect of ACE inhibition therefore remainsthe reduction in albuminuria. These results may therefore beLong-term studies in NIDDM. Effects on nephropathy and interpreted as supporting the major role of albuminuria in theretinopathy progression of renal disease. We have examined the protect-

Mordchai Ravid, Yosefa Bar-Dayan, Zohar Levi, David ive effect of ACE inhibitors also in the very-low-risk group ofBrosh, Dont Ravid-Safran, Rita Rachmani; Department of normotensive–normoalbuminuric patients with NIDDM.Medicine, Meir Hospital, Kfar-Sava and Sackler Faculty of Six-year follow-up on 156 patients, randomized to receiveMedicine, Tel-Aviv University, Israel enalapril or placebo showed a modest but a significant effect

of enalapril on attenuation of the decline in creatinine clear-The last decade of the 20th century has witnessed a markedance and on albuminuria. The mean decrease in creatinineincrease in the incidence of diabetes mellitus, mainly of theclearance was 2.5 ml/min/year in the placebo group and l.5non-insulin-dependent variety. This and the rise in life expect-m1/min/year in those who received enalapril (P=0.040). Theancy forecast a major load of end-stage renal disease on theinstitution of enalapril treatment was followed by an initialhealth-care systems of the industrial as well as some develop-decline in creatinine clearance of about 3 ml/min. Subsequenting countries. Effective policies to attenuate the progress offollow up showed stabilization throughout the trial period.diabetic nephropathy are therefore of increasing importance.The difference between the placebo and the treated arms,There is a general agreement about the impact of hyperten-therefore became significant only after 5 years. There was asion on the pace of deterioration in kidney function and ofsignificant association between the initial decline in UAE ratethe importance of albuminuria both as a prognostic para-by ACE inhibition and the subsequent renoprotective effect.meter and as a contributing factor to disease progression.15/79 placebo treated and 5/77 enalapril treated patientsThere is no consensus however about the desired levels ofcrossed the threshold to microalbuminuria (P=0.042).blood pressure or about the link between decrease in urinary

The presence of diabetic retinopathy was recorded by analbumin excretion (UAE) and the extend of renoprotection.annual ophthalmoscopic examination. In diabetics withAlso the relative efficacy of the various therapeutic regimesmicroalbuminuria, there were fewer new cases of retinopathyremains unsettled.among those treated with enalapril: (7.8 vs 19% among theACE inhibitors are advocated by some researchers as theplacebo group patients over a period of 5 years). Among thepreferred agents for renoprotection. Others, however, shownormoalbuminuric patients fewer new cases of retinopathysimilar effects on the kidney also of calcium channel blockingwere found in those treated with enalapril: (6%—new casesagents and of beta blocking drugs [1–3]. Meta-analysis of theamong enalapril-treated vs 17.8% in the placebo group duringvarious therapeutic trials is a frustrating task due to major6 years). When pulled together there were 2794 patient-yearsdiscrepancies in baseline characteristics of the patients especi-of follow-up. Enalapril treatment resulted in an absolute riskally in terms of risk factors (e.g control of hyperglycaemia,reduction of 11.5 percentage points for the development oflipid profile, age, duration of diabetes) and trial design. Someretinopathy (95% CI 6–27, P=0.004). There were very fewstudies include patients with and without nephropathy, withpatients with proliferative retinopathy and this developmentand without hypertension and one or both types of diabetes.seemed to be independent of therapy. The results of bothFurthermore, in studies of similar design the results are diver-trials indicate that the introduction of ACE inhibitor therapygent largely due to differences in surrogate end-points, choicemay have a long-term protective effect on the microvascularand dosage of therapeutic agents and duration of the studydisease of diabetes also in low-risk patients. In the kidney, the[4–6 ].main ACE inhibitor induced modification is the decline inLong-term studies in NIDDM have shown a varyingalbuminuria. The protection of the retinal blood vessels maydegree of decline in GFR in hypertensive and normotensivebe partially due to the low blood pressure and possibly alsopatients with nephropathy which was partially attenuated byby direct effects on the capillary wall.ACE inhibitors and by calcium blockers. The short-term

studies, lasting 1 year or less have often failed to demonstratesignificant change in kidney function, and therefore also no l. Velussi M, Brak E, Fragaede F et al. Effect of cilazapril and

amlodipine on kidney function in hypertensive NIDDM patients.effect of either class of agents on renal function. The onlyDiabetes 1996; 45: 216–222universally agreed effect is the decline in albuminuria induced

2. Chan JCN, Cockram CS, Nicholls MG et al. Comparison ofby ACE inhibitors although this effect was also observedenalapril and nifedipine in treating non-insulin dependent diabetesunder calcium blockers.associated with hypertension: one year analysis. Br Med J 1992;The studies we have performed are of 5 and 6 years dura- 305: 981–985

tion. The patient population was uniform and they were 3. Melbourne Diabetic Nephropathy Study Group. Comparisonplacebo controlled [7]. We therefore believe that our results between perindopril and nifedipine in hypertensive and normo-are reliable. They represent, however, highly selected groups tensive diabetic patients with microalbuminuria. Br Med J 1991;

302: 210–216of patients low and very low risk. The extrapolation of our4. Ruggenenti P, Moscone L, Bianchi L et al. Long-term treatmentresults to the diabetic population at large may therefore be

with either enalapril or nitrendipine stabilizes albuminuria andproblematic. The effect of a given therapeutic intervention isincreases glomerular filtration rate in non-insulin-dependent dia-expected to correlate to the degree of risk of the treatedbetic patients. Am J Kidney Dis 1994; 24: 753–761patients. The finding of a beneficial effect in a low-risk group

5. Lebovitz HE, Wiegmann TB, Cnaan A et al. Renal protectivemay therefore indicate the importance of such treatment in effects of enalapril ‘m hypertensive NIDDM: role of baselineother clinical settings. In the first study normotensive type 2 albuminuria. Kidney Int 1994; 45: S150-S155diabetics with microalbuminuria were randomized to receive 6. Bakris G, Barnhill B, Sadler R. Treatment of arterial hypertensionenalapril or placebo and were followed for 5 years. A clear in Diabetic humans: importance of therapeutic selection. Kidney

Int 1992; 41: 912–919renoprotective effect of ACE inhibition was demonstrated.

Abstracts1064

7. Ravid M, Savin H, Jutrin I et al. Long-term stabilizing effect of University Hospital, Aarhus Kommunehospital, Aarhus,angiotensin-converting enzyme inhibition on plasma creatinine Denmarkand on proteinuria in normotensive type II diabetic patients. Ann

Recent data indicate that substantial pathophysiologicalIntern Med 1993; 118: 577–581changes have taken already in the microalbuminuric stage.Blood pressure is elevated with an attenuated circadian

ACE inhibitors for protection against microvascular rhythm and vagal function is impaired with an abnormalcomplications in patients with type 1 diabetes sympathovagal interaction. Furthermore, new findings indi-

cate abnormalities in kidney ultrastructure in type 1 diabeticNishi Chaturvedi and the EUCLID Study Grouppatients with microalbuminuria. In the light of these abnor-Microvascular complications in people with type 1 diabetesmalities early intervention in microalbuminuria has acquiredcontinue to pose a major threat to health. ACE inhibitorsincreasing interest.slow the decline of renal function in advanced renal disease,

In two randomized placebo controlled double blind studiesbut their effects at earlier stages are unclear, and the optimalthe effect of 2 years treatment with either lisinopril treatmentlevel of albuminuria at which to institute treatment is not(20 mg) or placebo was evaluated in normotensive, micro-known. Further, whilst it is well known that retinopathyalbuminuric type 1 diabetic patients. In collaboration withshares many of the risk factors for nephropathy, the effects ofcolleagues in Padova, we here present a post hoc analysisACE inhibitors on retinopathy are not clear. We have there-of a subgroup of 60 patients with urinary albumin excre-fore examined the effects of the ACE inhibitor lisinopril ontion (UAE) between 20 and 70 mg/min. Baseline UAEnephropathy and retinopathy in ‘normotensive’ type 1was almost identical in the two groups (placebo, 36.3 mg/mindiabetic patients. EUCLID is a randomized, double-blind×/[1.4; lisinopril, 35.5 mg/min×/[1.5 (geometric meanplacebo controlled trial of the ACE inhibitor lisinopril in 530×/[tolerance factor)), whereas development in UAE overmen and women with type 1 diabetes aged 20–59 years withthe 2 years was significantly different (P<0.02) in thenormoalbuminuria or microalbuminuria. Patients weretwo groups with final UAE in the placebo group ofrecruited from 18 European centres, and were not on medica-58.8 mg/min×/[3.2 and 29.8 mg/min×/[2.5 in the lisinop-tion for hypertension. Resting blood pressure at entry wasril group. In the lisinopril group 22 patients (69%) revertedbetween 75 and 90 mmHg diastolic, and 155 mmHg systolic.to normoalbuminuria compared to 6 patients (21%) in theUrinary albumin excretion rate (AER) was centrally assessedplacebo group (P<0.01).by two overnight urine collections at baseline, 6, 12, 18 and

In the subgroup of patients examined in Aarhus (n=22) we24 months. Retinal photographs were taken at entry and 24performed 24-h ambulatory blood pressure measurementsmonths, and graded into 5 groups (‘none’ to ‘proliferative’).(AMBP), renal function tests (constant infusion technique)At 2 years, AER was 18.8% lower in the lisinopril comparedand determinations of exercise-induced albuminuria (bicycleto placebo group (95% CI 2.0, 32.7, P=0.03), adjusted for

baseline AER and centre; absolute difference 2.2 g/min. ergometer, 70% of estimated maximal VO2). AMBP showedIn people with normoalbuminuria, the treatment difference small increases in 24-h systolic and diastolic AMBP (1.6±6.2

was 12.7% (95% CI−2.9, 26.0, P=0.1), absolute difference and 0.7±4.9 mmHg over 2 years) in the placebo group, as1.0 g/min, whilst in those with microalbuminuria, this differ- opposed to significant reductions in the lisinopril groupence was 49.7% (95% CI−14.5, 77.9, P=0.1), absolute (−6.0±8.2 and −4.1±6.4 mmHg), (P<0.02 and 0.05difference 34.2 mg/min (P=0.04 for interaction). Retinopathy between the values)). Clinic BP measurements did not showprogressed by at least one stage in 13% of the lisinopril group significant differences. There were no differences in GFR orand in 24% of the placebo group (P=0.02). Lisinopril also RPF in the two groups, but development in UAE and devel-reduced progression to proliferative retinopathy, odds ratio opment in filtration fraction (FF) was positively correlated0.18 95% CI 0.04, 0.82, P=0.03. Treatment reduced retinopa- in the intervention group (r=0.9, P>0.01), i.e. the patientsthy incidence, odds ratio 0.69, 95% CI 0.30, 1.59, P=0.4. who showed the greatest fall in UAE also were the ones withLisinopril slows the progression of renal disease in normot- the greatest fall in FF. Exercise testing showed a numericallyensive type 1 diabetic patients with little or no albuminuria, reduced increase in exercise induced albuminuria in the lisino-but the greatest effect was observed in those with micro- pril group compared to the placebo group, although thealbuminuria (AER 20 g/min). Lisinopril also reduces the difference was not statistically significant.risk of progression of retinopathy. The use of ACE inhibitors In conclusion, ACE-i treatment in patients with low-gradeshould be considered in the early stages of type 1 diabetes. microalbuminuria reduces 24-h AMBP without attenuating

diurnal blood pressure variation, reduces UAE significantly,Chaturvedi N, Sjolie A-K, Stephenson JM, et al. Effect of lisinopril with changes in UAE being strongly associated with changes

on progression of retinopathy in normotensive people with type in FF. Furthermore, ACE-i reverses micro- to normo-1 diabetes. Lancet 1998; 351: 28–31.

albuminuria in a significant fraction of patients comparedThe Euclid study group. Randomised placebo-controlled trial ofto placebo.lisinopril in normotensive patients with insulin-dependent diabetes

and normoalbuminuria or microalbuminuria. Lancet 1997; 349:1787–1792 Mogensen CE, Keane WF, Bennett PH, Jerums G, Parving H-H,

Chaturvedi N, Fuller JH. Retinopathy in relation to albuminuria Passa P, Steffes MW, Striker GE, Viberti GC. Prevention of dia-and blood pressure in IDDM. In Mogensen CE (ed). The Kidney betic renal disease with special reference to microalbuminuria.and Hypertension in Diabetes Mellitus. Kluwer Academic Lancet 1995; 346: 1080–1084Publishers, Boston, 1996; 299–305 Hansen KW, Mau Pedersen M, Marshall SM, Christiansen JS,

Mogensen CE. Circadian variation of blood pressure in patientswith diabetic nephropathy. Diabetologia 1992; 35: 1074–1079

Poulsen PL, Hansen KW, Mogensen CE: Ambulatory blood pressureEarly intervention in microalbuminuria: 24-h BP and exercisein the transition from normo- to microalbuminuria; a longitudinalchanges study in IDDM patients. Diabetes 1994; 43: 1248–1253

Per L. Poulsen, Eva Ebbehøj, Carl Erik Mogensen; Medical Mølgaard H, Christensen PD, Hermansen K, Sørensen KE,Christensen CK, Mogensen CE. Early recognition of autonomicDepartment M, Diabetes & Endocrinology, Aarhus

Abstracts 1065

dysfunction in microalbuminuria: significance for cardiovascular similar decline in AER and GFR was observed in patientsmortality in diabetes mellitus? Diabetologia 1994; 37: 788–796 treated with cilazapril or amlodipine [8]. The rate of GFR

Østerby R, Bangstad HJ, Nyberg G, Walker JD, Viberti GC. A fall in the study group as a whole was inversely related to thequantitative ultrastructural study of juxta glomerular arterioles in decrease in mean blood pressure, suggesting that arterialIDDM patients with micro- and normoalbuminuria. Diabetologia hypertension modulates the progression of renal damage,1995; 38: 1320–1327

even when AER is normal. A 12 month comparison of enala-Bangstad HJ, Østerby R, Dahl Jørgensen K, Berg KJ, Hartmann A,pril and nitrendipine in 16 microalbuminuric type 2 diabeticNyberg G, Bjørn SF, Hanssen KF. Early glomerulopathy is pre-patients with biopsy-proven DN showed that AER wassent in young, type 1 (insulin-dependent) diabetic patients withstabilized and GFR increased in both groups [9]. After 27microalbuminuria. Diabetologia 1993; 36: 523–529

Poulsen PL, Ebbehøj E, Hansen KW, Mogensen CE. 24-h blood months, AER decreased in both groups compared to baselinepressure and autonomic function is related to albumin excretion and GFR remained at or above baseline levels in both groups.within the normoalbuminuric range in IDDM patients. A recent study in type 2 diabetic patients examined theDiabetologia 1997; 40: 718–725 hypothesis that ACEi and CCB mediated reductions in pro-

The Microalbuminuria Captopril Study Group. Captopril reduces teinuria result in equal slowing of progression of DN, giventhe risk of nephropathy in IDDM patients with microalbuminuria. similar levels of blood pressure control [10]. Type 2 diabeticDiabetologia 1996; 39: 587–593

patients with hypertension and overt DN were treated withThe Euclid Study Group. Randomised placebo-controlled trial ofeither lisinopril (n=18), verapamil or diltiazem (n=18) orlisinopril in normotensive patients with insulin-dependent diabetesatenolol (n=16), with no significant difference in reductionand normoalbuminuria or microalbuminuria. Lancet 1997; 349:in mean blood pressure over 63 months. Atenolol-treated1787–1792patients had a greater rate of decline in creatinine clearanceand a lesser fall in AER. However, no differences were dis-cerned between the ACEi and CCB groups. In a 42 monthstudy from Denmark, hypertensive type 2 diabetic patients

Ace inhibition vs calcium-channel blockade in normotensive with DN were treated with lisinopril (n=21) or atenolol (n=type 1 and type 2 diabetic patients with microalbuminuria 22), in equihypotensive doses. After 6 months, GFR declined

by less than 1 ml/min/month in both groups suggesting thatG. Jerums on behalf of the Melbourne Diabetic Nephropathythe progressive decline in GFR in DN can be amelioratedStudy Group, Melbourne, Australiaequally effectively by both treatments even though AER was

Several studies have examined whether angiotensin con- reduced more in the lisinopril than the atenolol group [11].verting enzyme inhibitors (ACEi) exert beneficial effects on In 1991, the Melbourne Diabetic Nephropathy Studythe course of diabetic nephropathy (DN) which extend Group reported the results of 12 months of treatment withbeyond blood pressure control. In type 1 diabetes, such an equihypotensive doses of perindopril or nifedipine (sloweffect has been demonstrated in patients with advanced DN release) in type 1 and type 2 diabetic patients with microalbu-[1] and the antiproteinuric effect of ACEi has been linked to minuria [12]. Both treatments reduced AER in hypertensivesubsequent protection from a decline in glomerular filtration patients and stabilised AER in normotensive patients. Arate (GFR) [2]. However, in normotensive microalbuminuric second study is now in progress, with a follow-up ofpatients it has been difficult to determine if the initial decline 2–8 years, in normotensive microalbuminuric (AERin albumin excretion rate (AER) is linked to protection from 20–200 g/min in 2 out of 3 measurements) patients with typethe subsequent decline in GFR, even after a follow-up of 8 1 or type 2 diabetes. Entry systolic blood pressure (SBP) wasyears [3]. <140 mmHg if age <40 years, otherwise <160 mmHg and

The renal effects of ACEi in overt DN have been compared diastolic blood pressure (DBP) <90 mmHg. Eighty-onewith those of other antihypertensive agents by meta-analysis. patients were randomly assigned to receive placebo (PLAC ),ACEi were shown to decrease proteinuria independently of perindopril (PER) or nifedipine (NIF) using an open labelchanges in blood pressure and had an additional favourable protocol. Perindopril 2–8 mg/day and nifedipine (sloweffect on GFR that was independent of changes in blood release) 10–40 mg/day were titrated to achieve a fall in DBPpressure. ACEi and calcium-channel blockers (CCB) other >4 mmHg. In each participant, mean SBP, DBP and MBPthan nifedipine had very similar renal protective effects, during the study, and gradients in AER (% change/year), andexpressed as changes in proteinuria or GFR, whereas nifedip- calculated GFR (Cockcroft–Gault, ml/min/year), were calcu-ine therapy did not reduce proteinuria and was associated lated from 3 monthly measurements. The development ofwith a more rapid decline in GFR [4]. macroalbuminuria was noted. In addition, yearly measure-

It is not clear to what extent studies in type 1 diabetes can ments of GFR using 99mTechnetium-DTPA and 24-hbe translated to type 2 diabetic patients. In type 2 diabetes, ambulatory blood pressure were performed.ACEi has been shown to reduce AER and to present a decline In both type 1 and type 2 diabetic patients, HbA1c levelsin GFR as estimated on the basis of the reciprocal of plasma were not significantly different in any treatment group.creatinine levels, when studied over 7 years in middle-aged In the 47 type 2 diabetic patients, mean clinic arterial pres-Israeli subjects [5]. A similar study performed over 4 years in sure during the study was similar in the 3 groups (PER 98,normotensive Japanese type 2 diabetic patients, showed that NIF 97, PLAC 100 mmHg). However, mean 24-h ambulat-ACEi reduced AER but did not change creatinine clearance ory blood pressure was lower in perindopril treated patients[6 ]. (PER 90, NIF 100, PLAC 97 mmHg, PER vs NIF P<0.05).

Several recent studies have directly compared CCB with Baseline AER levels were: PER 96, NIF 60, PLAC 86 g/min,ACEi in evolving DN. In normotensive type 1 microalbumin- n.s. Individual AER gradients showed no tendency touric patients, a placebo-controlled study compared the effects increase in any treatment group (PER−4, NIF −1, PLACof lisinopril and nifedipine over 4–10 years [7]. Both drugs +4%/year, n.s.). Eleven patients developed macroalbuminu-were equally effective in delaying the occurrence of overt pro- ria (PER 2, NIF 3, PLAC 6, n.s.). Mean baseline GFR wasteinuria, despite lower systolic blood pressure levels in the similar in all three groups (PER 116, NIF 110, PLAClisinopril-treated group. In a 3-year study of 44 hypertensive 119 ml/min/1.73 m2, n.s.). Individual GFR gradients in the

placebo group showed a significant downward trendtype 2 diabetic patients with normo- or microalbuminuria, a

Abstracts1066

ensive type 2 diabetic patients: results of a 4-year, prospective,(7.5±2.7 ml/min/year) but were stable in the other groupsrandomized study. Diab Med 1996; 13: 120–124(PER +1.1, NIF −0.6 ml/min/year). A specific treatment

7. Crepaldi G, Carta Q, Deferrari G et al. Effects of lisinopril andeffect on measured or calculated GFR was not evident.nifedipine on the progression to overt proteinuria in normotens-In 34 type 1 diabetic patients, mean clinic blood pressureive insulin dependent diabetic patients with incipient nephro-during the study was similar in each group (PER 92, NIF 96, pathy. Diabetologia 1997; 40: suppl. 1: A545

PLAC 96 mmHg, n.s.), but mean 24-h ambulatory blood 8. Velussi M, Brocco E, Frogato F et al.. Effects of cilazapril andpressure was lower in perindopril treated patients (PER 88, amlodipine on kidney function in hypertensive NIDDM patients.NIF 92, PLAC 95 mmHg, PER vs PLAC P<0.05). Mean Diabetes 1996; 45: 216–222

9. Ruggenenti P, Mosconi L, Bianchi L et al.. Long-term treatmentbaseline AER was: PER 67, NIF 75, PLAC 78 g/min, n.s.with either enalapril or nitrendipine stabilizes albuminuria andIndividual AER gradients showed a significant increase inincreases glomerular filtration rate in non-insulin-dependentboth the nifedipine and placebo groups, but remained stablediabetic patients. Am J Kidney Dis 1994; 24: 753–761in the perindopril group (PER −4, NIF 49, PLAC 20%/year,

10. Bakris GL, Copley JB, Vicknair N, Sadler R, Leurgans S.NIF vs PER P=0.02, NIF vs PLAC P=0.05). ElevenCalcium channel blockers versus other antihypertensive therapiespatients developed macroalbuminuria (PER 1, NIF 5, PLAC on progression of NIDDM associated nephropathy. Kidney Int

5, PER vs NIF P=0.05, PER vs PLAC P=0.06). The 1996; 50: 1641–1650mean baseline GFR (PER 118, NIF 132, PLAC 11. Nielsen FP, Rossing P, Gall M-A, Skott P, Smidt UM, Parving120 ml/min/1.73m2, n.s.) showed a trend to a higher baseline H-H. Long-term effect of lisinopril and atenolol on kidney

function in hypertensive NIDDM subjects with diabetic nephro-GFR in the nifedipine group. This was associated with anpathy. Diabetes 1997; 46: 1182–1188accelerated decline in GFR in the nifedipine group (PER −3,

12. Melbourne Diabetic Nephropathy Study Group. ComparisonNIF −16, PLAC 0 ml/min/year, NIF vs PLAC iP<0.01,between perindopril and nifedipine in hypertensive and normo-NIF vs PER P<0.05). However, attained GFR at 4 yearstensive diabetic patients with microalbuminuria. Br Med J 1991;approximated to 100 ml/min/1.73m2 in all three treatment302: 210–216groups. Calculated GFR was generally lower than measured

GFR and declined by approximately 3 ml/min/year, with noevidence of an accelerated rate of decline in the nifedipinegroup.

In summary, no evidence of progression of AER or GFR Long term follow-up of the glomerular filtration rate inwas noted in microalbuminuric type 2 diabetic patients who normotensive type 1 diabetic subjects with microalbuminuriaremained normotensive or whose blood pressure was main- during angiotensin 1 converting enzyme inhibitiontained below 160/90 after a mean follow-up of 4 years. No

Michel Marre, Patrick Fabbri, Beatrice Bouhanick, Gillesspecific renoprotective effect of perindopril or nifedipineBerrut, Jean-Jacques Le Jeune, Francoise Bled; Centrecould be demonstrated. By contrast, in type 1 diabeticHospitalier Universitaire, 49033 Angers Cedex 01, Francepatients, AER did increase significantly in the placebo and

nifedipine groups, and this was prevented by perindopril. Objectives. The glomerular filtration rate can be protected byDefinitive assessment of the effects of nifedipine and perindo- angiotensin I converting enzyme inhibition in subjects whopril on progression of GFR will require longer-term study as develop diabetic nephropathy. Diabetic nephropathy (indi-well as further assessments of GFR by: (a) calculation of cated by proteinuria, or macroalbuminuria, >300 mg/24 h)GFR gradients weighted for duration of therapy to allow for can be prevented by angiotensin I converting enzyme inhibi-unequal follow up; (b) calculation of GFR gradients with tion in normotensive type 1 diabetic subjects with micro-and without baseline values to allow for initial treatment albuminuria (30–300 mg/24 h). However, the glomerular fil-effects on GFR; (c) measurement of GFR after cessation of tration rate is normal or supranormal in these patients. Wedrug therapy. therefore tested the value of early angiotensin I converting

If an effect of either drug on GFR progression is found, it enzyme inhibition to maintain the glomerular filtrationwill then be necessary to determine if this can be separated rate of normotensive type 1 diabetic subjects with micro-from an effect on systemic blood pressure. Current evidence albuminuria in the long term.suggests that clinic blood pressures may be an unreliable Design. Prospective, open, sequential, follow-up study of theindicator of 24-h ambulatory blood pressure levels in normot- glomerular filtration rate in normotensive type 1 diabetic sub-ensive patients. jects with microalbuminuria receiving one of two treatment

strategies with an angiotensin I converting enzyme inhibitor1. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect of (enalapril 20 mg/day): early, from microalbuminuria identi-

angiotensin converting enzyme inhibition on diabetic nephro- fication (strategy 1), or late, only if patients developedpathy. N Engl J Med 1993; 329: 1456–1462 macroalbuminuria (strategy 2).

2. Wapstra FH, Navis G, de Jong PE, de Zeeuw D. Prognostic Setting. Diabetic clinic in a tertiary referral centre.value of the short-term antiproteinuric response to ACE inhibi- Subjects. Twenty-five normotensive (blood pressuretion for prediction of GFR decline in patients with nondiabetic

<160/95 mmHg) type 1 diabetic subjects with micro-renal disease. Exp Nephrol 1996; 1: 47–52albuminuria.3. Mathiesen E, Hommel E, Smith U, Parving H-H. Efficacy ofInterventions. In 15 subjects enalapril was given immediatelycaptopril in normotensive diabetic patients with microalbuminu-

ria–8 years followup. Diabetologia 1995; 38: suppl. 1: A46 once microalbuminuria was identified (strategy 1), and in 104. Weidmann P, Schneider M, Bohlen L. Therapeutic efficacy of others only if they progressed to macroalbuminuria (strategy

different antihypertensive drugs in human diabetic nephropathy: 2). The blood-pressure target was BP <140/85 mmHg withan updated meta-analysis. Nephrol Dial Transpl 1995; 10: antihypertensive drugs other than angiotensin I convertingsuppl: 39–45 enzyme inhibitors.

5. Ravid M, Lang R, Rachmani R, Lishner M. Long-term renopro- Main outcome measures. Yearly assessment of glomerulartective effect of angiotensin-converting enzyme inhibition in non-filtration rate (using 125I-iodothalamate constant infusioninsulin-dependent diabetes mellitus. A 7-year follow-up study.technique), and classification as glomerular hyperfiltrationArch Intern Med 1996; 156: 286–289(>150 ml/min/1.73 m2, the mean plus 2 SD of 36 controls),6. Sano T, Hotta N, Kawamura T et al. Effects of long-term

enalapril treatment on persistent microalbuminuria in normot- normofiltration (99–150 ml/min/1.73 m2), or hypofiltration

Abstracts 1067

Table 1. Glomerulopathy data, mean and (CV%)(<99 ml/min/1.73 m2, the mean minus 2 SD of controls). Theprimary outcome measure was the number of subjects whoprogressed to glomerular hypofiltration. The rate of change ND Type 1 diabetics Pin the glomerular filtration rate (final minus initial divided byfollow-up duration) was also calculated. Urinary albumin

BMT 350 586 <10−4excretion and its status (normoalbuminuria <30 mg/24 h, nm (15) (17)microalbuminuria, macroalbuminuria) and blood pressure VV (mes/glom) 0.19 0.21 0.002were assessed twice a year. (9) (19)Results. The follow-up was 48 (range 24–72) months. At VV (matrix/glom) 0.09 0.12 <10−4

(16) (21)baseline, 6 subjects on strategy 1 had glomerular hyper-Matrix star volume 14 28 <10−4filtration, and 9 normofiltration, whereas 3 subjects onmm3 (33) (27)strategy 2 had hyperfiltration, and 7 normofiltration (n.s.). At

follow-up, 8 subjects on strategy 1 had hyperfiltration, and 7normofiltration, whereas one subject on strategy 2 had hyper-filtration, 4 normofiltration, and 5 hypofiltration (x2=10.69; volume fractions but the results in terms of change from firstP<0.01). On strategy 1 the median rate of change in the glo- to second biopsy were quite similar by EM and LM.merular filtration rate was 0.03 vs −0.86 ml/min/1.73 m2/ In a subset of the patients (six pair-biopsies) a comparisonmonth of follow-up on strategy 2 (Mann–Whitney U test of results was made using either three or five glomeruli. TheP=0.0016). These results were attributable to the enalapril conclusion concerning paired biopsies did not change withtreatment strategy but not to changes in blood pressure, or to the larger number studied.final status of urinary albumin excretion. In the baseline biopsies the type 1 diabetic group of patientsConclusions. Treatment with enalapril (20 mg/24 h) prevents had significantly increased structural parameters comparedthe glomerular filtration rate declining in normotensive type with the control group (see Table 1). Each of the parameters1 diabetic subjects with microalbuminuria. showed a positive correlation with the level of AER: index of

glomerulopathy vs AER r=0.47, P=0.003. The degree ofMarre M, Fabbri M, Berrut G., Bouhanick B. The concept of structural changes was moderate in all cases. Thus, these

incipient diabetic nephropathy and effect of early antihypertensive results conflict to some extent with those reported by Fiorettointervention. In: Mogensen CE (ed). The Kidney and Hypertension et al. [2], who found marked changes in several cases ofin Diabetes Mellitus. Kluwer, Boston, 1996; 351–360 normo- and low-grade albuminuria. This discrepancy can

probably be ascribed to the fact that all of our patients wereyoung with microalbuminuria appearing after 10–15 years ofdiabetes and none had hypertension. It seems that when mic-roalbuminuria appears after longer duration of diabetes andStructural changes in microalbuminuria, effect of interventionpatients’ age thus is higher, a different pattern of the

Ruth Østerby, Hans-Jacob Bangstad, Susanne Rudberg; structural changes is seen [3], probably dominated byElectron Microscopy Laboratory, Aarhus Kommune- macroangiopathy.hospitalet, Aarhus University Hospital, Aarhus, Denmark; The study dealing with the effect of intensified diabetesAker Diabetes Research Centre, Aker University Hospital, treatment [4] showed positive correlation between meanOslo, Norway; Pediatric Unit, Karolinska Instituttet, HbA1c during the study and increase in basement membraneStockholm, Sweden thickness (BMT) (r=0.70, P=0.001). Thus, although

optimal diabetes control was not achieved in these adoles-Two follow-up biopsy series have been obtained in microal-buminuric type 1 diabetic patients with the aim of testing the cents, the patients with the best control were better off in

terms of showing less progression of BM thickening.influence of intervention on renal structural changes. Thetreatment given during the interval between first and second The patients in this series have now been followed for 6 years

after the first kidney biopsy. Follow-up study with determina-biopsy was either intensified diabetes control vs conventionalcontrol (Norway, n=20) or ACE inhibitors vs beta-blockers tion of AER, HbA1c and blood pressure measurements have

shown that BMT, HbA1c and systolic blood pressure contrib-(Sweden, n=18). The analysis of follow-up biopsies has beencompleted only in the Norwegian series at the present time. uted significantly to the variation in AER at 6 years.

In the Swedish series the relationship between clinical vari-Both series dealt with young (age, median and (range): 19(14–29) years) patients with diabetes duration 12 (6–18) years ables during the years preceding the biopsy and the structural

parameters was studied. It was found that 5-year meanand AER 33 (15–195) mg/min, i.e. the majority of patients hadlow-range albuminuria. No patient had elevated blood pres- HbA1c, diabetes duration and GFR were the variables with

an independent influence on the severity of the diabetic glo-sure at entry into the study. Non-diabetic kidney donors (ND)of age <40 years constituted the reference group. merulopathy [5].

In this series a comparison was also made with the courseSince the aim was to compare change in structural dataover a fairly short period of time (2–3 years) it was mandat- of GFR, estimated by determinations during the normoalbu-

minuric stage 2–5 years prior to the renal biopsy and again atory to obtain the parameters with the highest possible preci-sion. Therefore some effort has been made to optimize the the time of the biopsy. Patients showing a decline in GFR

>5 ml /min/year had the thickest BM and matrix expansion.methodology. Estimation of mesangial volume fraction isespecially demanding since the changes are only borderline at In multiple regression analysis of structure vs clinical vari-

ables the sole variable independently associated with the fallthe early stage of diabetic nephropathy. Therefore, threelevels were analysed in each of three glomeruli as the standard in GFR was BMT (P=0.003) [6 ].

Thus, new little bits of information underline theprocedure by electron microscopy (EM ) [1]. Further, in oneof the series the measurements have been supplemented with association between diabetes control and the development of

structural changes in the kidney and between structure andquantitation by light-microscopy (LM ) of 1-mm-thick sec-tions, using the largest cross-sectional area of seven glomeruli function. The final analysis of the follow-up biopsies will

show if treatment with ACE-inhibitors for 3 years has anyin each case. Light-microscopy as expected led to higher

Abstracts1068

demonstrable effect on structural changes. Ongoing studies The concept of angiotensin II receptor blockers in hypertensionand renal diseaseshould further clarify if it holds true that morphometric stud-

ies are useful in intervention trials over short periods of time. Michel Burnier; Division of Hypertension and VascularMedicine, Lausanne, SwitzerlandBlockade of the renin–angiotensin–aldosterone system withangiotensin-converting enzyme (ACE) inhibitors is now1. Østerby R. Research methodologies related to renal complica-

tions: structural changes. In: Mogensen CE, Standl E, ed. recognized as an effective therapeutic approach to slow theResearch Methodologies in Human Diabetes, vol.2. Diabetes progression of chronic renal failure in diabetic and non-Forum Series V, Part II. Walter de Gruyter, Berlin, 1995; 289–309 diabetic nephropathies. Recently, non-peptide angiotensin II

2. Fioretto P, Steffes MW, Mauer SM. Glomerular structure in receptor antagonists have become available which enable tononproteinuric IDDM patients with various levels of albuminuria. inhibit the renin–angiotensin system by blocking the bindingDiabetes 1994; 43: 1358–1364 of angiotensin II to its AT1 receptor, this latter being respons-3. Østerby R, Schmitz A, Nyberg G, Asplund J. Renal structural

ible for all the known effects of angiotensin II. The renalchanges in insulin dependent diabetic patients with albuminuria.effects of these antagonists have now been characterized inComparison of cases with onset of albuminuria after short ornormal subjects and hypertensive patients with or withoutlong duration. APMIS (in press)renal failure. The results of these studies suggest that angio-4. Bangstad H-J, Østerby R, Dahl-Jørgensen K, Berg KJ, Hartmanntensin II receptor antagonists and ACE inhibitors have ratherK, Hanssen KF. Improvement of blood glucose control retards

the progression of morphological changes in early diabetic similar renal effects. Indeed, angiotensin II antagonists havenephropathy. Diabetologia 1994; 37: 483–490 been shown to increase renal plasma flow without affecting

5. Rudberg S, Østerby R, Dahlquist G, Nyberg G, Persson B. glomerular filtration rate; hence filtration fraction decreases.Predictors of renal morphological changes in the early stage of Significant increases in urinary sodium excretion were alsomicroalbuminuria in adolescents with IDDM. Diabetes Care observed during angiotensin II receptor blockade. In contrast1997; 20: 265–271 to other angiotensin II antagonists, losartan has a uricosuric6. Rudberg S, Østerby R. Decreasing glomerular filtration rate—

effect which may be useful in patients with hypertension andan indicator of more advanced diabetic glomerulopathy in thehyperuricaemia. In hypertensive patients, the uricosuric effectearly course of microalbuminuria in IDDM adolescents? Nephrolof losartan has been shown to prevent the diuretic-inducedDial Transplant 1997; 12: 1282–1283increase in plasma uric acid. Recently we have investigatedthe possible renal interaction between angiotensin II receptorantagonists and non-steroidal anti-inflammatory drugs(NSAID). Interestingly, although angiotensin II antagonists

Renal interstitial expansion is modified by perindopril in true do not affect prostaglandin metabolism, we found that indo-methacin blunts both the natriuretic response to the angio-diabetic glomerulosclerosis (DG in type 2 (NIDDM ) diabetictensin II antagonist valsartan and to ACE inhibitor enalapril.patients. A 2-year sequential biopsy studyThus NSAID may also attenuate the blood pressure loweringDaniel J. Cordonnier, N. Pinel, C. Barro, Philippe Zaoui, oneffect of angiotensin II receptor antagonists. Today thebehalf of the DIABIOPSIES GROUP; CHU Grenoble,important remaining question concerns the renal protectiveFranceeffects of angiotensin II receptor antagonists. Experimentally,angiotensin II blockers appear to have the same effect as ACEAngiotensin-converting enzyme inhibitors (ACE-I) exert ainhibitors, but the final clinical response will be provided byprotective effect against the progression of renal insufficiencythe results of the renal study.in type 1 diabetes. Our objective was to study the structural

effect of an ACEI, perindopril (PI ) in type 2 diabetic patientsBurnier M, Rutschmann B, Nussberger J, Versaggi J, Shahinfar S,with diabetic glomerulosclerosis (DG), two biopsies being

Waeber B, Brunner HR. Salt dependent renal effects of anperformed at initiation of the study and 2 years later. 22angiotensin II antagonist in healthy subjects. Hypertension 1993;patients were randomized, 11 to PI and 11 to placebo (PO), 22: 339–347

after informed consent. At baseline (MO), 18 were macro and Burnier M, Hagman M, Nussberger J, Biollaz J, Armagnac C,4 were microalbuminuric; diagnosis of DG (6 ‘early’ and 16 Brouard R, Waeber B, Brunner HR. Short-term and sustained‘typical’) was assessed by two pathologists; four pts were not renal effects of angiotensin II receptor blockade in healthy sub-

jects. Hypertension 1995; 25: 602–609included due to predominant ischaemic lesions. 1 patientBurnier M, Roch-Ramel F, Brunner HR. Renal effects of angiotensindecided to leave the study at M 21. Compliance was assessed

II receptor blockade in normotensive subjects. Kidney Int 1996;by plasma ACE activity. Tolerance was good. BMI and49: 1787–1790HbA1c did not change significantly. Blood pressure was

Gansevoort RT, De Zeeuw D, De Jong PE. Is the antiproteinuricslightly but not significantly more elevated in PO group. 24-h effect of ACE inhibition mediated by interference in the renin-albuminuria increased in PO group and diminished in PI angiotensin system? Kidney Int 1994; 45: 861–867group at 2 years (P<0.05). The relative surface of interstit- Soffer BA, Wright JT, Pratt JH, Wiens B, Goldberg AI, Sweet CS.

Effects of losartan on background of hydrochlorothiazide inium in the biopsy (stereological analysis/BIOCOM computer-patients with hypertension. Hypertension 1995; 26: 112–117ized method) increased in PO (31.7%±5.3 to 40.5±11.6) and

remained stable in PI (33.8%±4.8 to 34.6±6.6) (P<0.05).The relative volume of mesangium per glomerulus showedthe same trend without reaching significance. Collagen VI, amarker of DG in man, was increased further in interstitium

A follow-up study of the course of nephropathy in type 1of PO than in those of PI (P<0.01); the same was observeddiabeteswith thrombospondin (P<0.01).Ed Lewis; Section of Nephrology, Rush Presbyterian St,Conclusion. Despite limited patient population and observa-Lukes Medical Center, Chicago, USAtion time, renal biopsies advocate in this study for the role of

angiotensin in interstitial expansion, a major factor of renal The purpose of the study was to determine the clinical courseof type 1 diabetic nephropathy in patients randomized to twofibrosis.

Abstracts 1069

different levels of blood pressure control. 129 patients previ- Subjects and methodsously entered into our ACE inhibitor trial were randomized The analysis was performed in a subset of patients (n=35, 22to intensive BP control (MAP 92 mmhg) or standard BP M/13 F) who had microalbuminuria (10–100 mg/12 h in ancontrol (MAP 100–107 mmHg). overnight urine sample) at randomization in a large double-

Baseline characteristics included 47% male; 97% white; age blind, placebo-controlled study, investigating the effect of37 years; age of onset of type 1 DM 11 years; duration type 1 candesartan on blood glucose homeostasis and blood lipidDM 26 years; duration of proteinuria 7.4 years; duration of profile in patients with stable type 2 diabetes (aged 30–75retinopathy 9.3 years; mean serum creatinine 1.7 mg/dl years, HbA1c 5.5–9.0%), mild hypertension (sitting diastolic(150 pmol/l ); mean proteinuria 2.0 g/24h; iothalamate–GFR blood pressure 90–100 mmHg) and serum creatinine below66 ml/min/1.73; glycosylated Hb 11.6%. Median dose of ram- 150 mol/l for men and 120 mol/l for women. Patients wereipril in the standard group was 2.5 mg/day and the intensive randomized, after a 4-week placebo run-in period, to 12group 10 mg/day. Median MAP achieved for the standard weeks double-blind treatment with candesartan 8–16 mg (n=group was 97 mmHg and the intensive group 90 mmHg. 15) or placebo (n=20) once daily. The differences betweenMedian 24-h creatinine clearance in the standard group was treatments in change in efficacy variables from randomization70 ml/min at baseline and 65 ml/min at 24 months follow-up. to the end of the study were analysed.Median creatinine clearance for the intensive group was55.5 ml/min at baseline and 53.6 ml/min at 24 months follow-

Resultsup. Median 24-h proteinuria was 1140 at baseline and1723 mg/24 h at 24 months follow-up for the standard group Candesartan did not influence blood glucose homeostasis orand for the intensive group it was 1044 at baseline and blood lipid profile compared to placebo. The body weight535 mg/24 h at 24 months follow-up. remained unchanged in both treatment groups. The

During the follow-up period the standard group experi- median urinary albumin excretion decreased from 28.5 toenced one serum creatinine doubling and 6 ESRD as com- 12.2 mg/12 h (−57%) in patients treated with candesartanpared to the intensive group with one doubling and 3 ESRD. (n=15) while it increased from 30.2 to 32.8 mg/12 h (+9%)

We conclude: (1) ramipril is a renoprotective ACE inhib- in the placebo group (n=20, P=0.03 for the differenceitor; (2) the combination of ramipril at higher dose level plus between treatments). In this subset of patients the meanMAP 92 mmHg is associated with regression of clinical reduction in the diastolic blood pressure was 6.4 mmHg innephropathy; (3) half of the patients in the intensive BP con- the candesartan group and 3.6 mmHg in the placebo grouptrol group no longer meet criteria for overt clinical (P=NS).nephropathy.

ConclusionsBjorck S. Clinical trials in overt diabetic nephropathy. In: MogensenCE (ed). The Kidney and Hypertension in Diabetes Mellitus. In this placebo-controlled study, 12 weeks of treatment withKluwer, Boston 1996; 375–382 the angiotensin II antagonist candesartan reduced microalbu-

minuria in patients with stable type 2 diabetes and mildhypertension. Thus, candesartan appears to have a nephro-protective potential in this patients category. Further studiesare needed to evaluate the long-term beneficial effects of thisAngiotensin II receptor blockers in type 2 diabetic patients new class of drugs.with microalbuminuria

Carol Forsblom, P. Trenkwalder, K. Dahl, H. Mulder, on 1. Viberti GC, Hill RD, Jarrett RJ, Argyropoulos A, Mahmud U,Keen H. Microalbuminuria as a predictor of clinical nephropathybehalf of the Multicentre Study Group (Finland, Germany,in insulin-dependent diabetes mellitus. Lancet 1982;1: 1430–1432Norway, and the Netherlands); Helsinki University Hospital,

2. Mogensen CE. Microalbuminuria predicts clinical proteinuriaDepartment of Medicine, Division of Internal Medicine,and early mortality in maturity onset diabetes. N Engl J MedHelsinki, Finland1984; 310: 356–360

3. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD, The CollaborativeStudy Group. The effect of angiotensin-converting-enzyme inhibi-Introduction tion on diabetic nephropathy. N Engl J Med 1993; 329: 1456–1462

Microalbuminuria is a predictor of nephropathy in patients 4. Parving HH, Rossing P, Hommel E, Smidt UM. Angiotensin-converting enzyme inhibition in diabetic nephropathy: ten years’with type 1 or type 2 diabetes mellitus [1,2] and interventionsexperience. Am J Kidney Dis 1995; 26: 99–107that decrease albuminuria are likely to postpone the develop-

5. Gansevoort RT, de Zeeuw D, Shahinfar S, Redfield A, de Jongment of severe renal impairment. Blocking the renin–angiot-PE. Effects of the angiotensin II antagonist losartan in hypertens-ensin system via ACE inhibition has been effective in thisive patients with renal disease. J Hypertens 1994; 12: [Suppl.respect [3,4]. The more direct route of inhibiting the negative2]: S37–S42

effects of angiotensin II by means of an angiotensin II recep- 6. Goldberg AI, Dunlay MC, Sweet CS. Safety and tolerability oftor antagonist is expected to provide at least similar reduc- losartan potassium, an angiotensin II receptor antagonist,tions of albuminuria [5]. This new class of drugs offer compared with hydrochlorthiazide, atenolol, felodipine ER, andtherapeutic advantages like a reduction in the incidence of angiotensin-converting enzyme inhibitors for the treatment of

systemic hypertension. Am J Cardiol 1995; 75: 793–795cough and inflammation, seen with ACE inhibitors [6 ].Candesartan is a new angiotensin II type 1 receptor antagon-ist with tight binding to and a slow dissociation from thereceptor.

Advanced glycation end products and inhibitorsMark E. Cooper; Department of Medicine, University of

Aim Melbourne, Austin & Repatriation Medical Centre, WestHeidelberg 3081, AustraliaTo assess the effect of candesartan on microalbuminuria in

patients with type 2 diabetes and mild hypertension. The process of advanced glycation involves a spontaneous

Abstracts1070

of p60 to OST–48 and p90 to 80K-H membrane proteins. Procreaction between glucose and proteins and lipids, particularlyNatl Acad Sci USA 1996; 93: 11047–11052on long-lived structural proteins such as the collagens, leading

10. Kochakian M, Manjula B, Egan J. Chronic dosing with aminog-to the formation of advanced glycation end products (AGEs)uanidine and novel advanced glycosylation end product forma-[1]. This process not only involves glycation but also oxida-tion inhibitors ameliorates cross-linking of tail tendon collagention steps, and therefore many investigators now consider in streptozotocin-induced diabetic rats. Diabetes 1996; 45:

these products to be as a result of glucoxidation [2]. A range 1694–1700of these products have been detected both in vitro and in vivo, 11. Vasan S, Zhang X, Zhang X, Kapurniotu A, Bernhagen J,some of these AGEs having a characteristic fluorescence [3]. Teichberg S, Basgen J, Wagle D, Shih D, Terlecky I, Bucala R,

An inhibitor of this pathway, aminoguanidine, has been Cerami A, Egan J, Urlich P. An agent cleaving glucose-derivedprotein cross-links in vitro and in vivo. Nature 1996; 382: 275–278shown to decrease AGE formation in rat tissues including the

aorta [3] and the kidney [4]. This treatment is associated withattenuation in the rise in albuminuria and prevention of mes-angial expansion in the diabetic rat [4]. Similar beneficialeffects on other diabetic vascular complications have been

Cytokines and growth factors in diabetic renal involvement:reported and recent studies by our group have suggested thatfocus on transforming growth factor-bthe protective actions of aminoguanidine involve prevention

of overexpression of prosclerotic cytokines such as trans- Kumar Sharma, Fuad N. Ziyadeh; Thomas Jeffersonforming growth factor beta (TGF-b1) and matrix proteins University; University of Pennsylvania, Philadelphia, PA,such as type IV collagen [5]. Aminoguanidine has now been USAadministered to man and shown to inhibit haemoglobin AGE

A variety of growth factors and vasoactive peptides have beenlevels [6 ]. This drug is at present under trial in patients withimplicated in the pathogenesis of diabetic kidney disease.end-stage renal disease and in diabetic patients with microal-Growth hormone and the related insulin-like growth factorsbuminuria and overt nephropathy. The results of thesehave been shown to play roles in diabetic renal hypertrophyclinical trials are awaited with great interest.and in mesangial matrix accumulation in experimentalA range of receptors for AGEs have been isolated andanimals [1], tumour necrosis factor-a and PDGF-BB haveshown to be present in the kidney [7–9]. The effects of dia-been found to be elevated in glomeruli of diabetic rats [2],betes and various AGE inhibitors on expression of theseand the vasoactive peptides endothelin [3] and angiotensin IIreceptors is an area of intense investigation.[4] are also involved in diabetic renal disease. However, theRecently more potent inhibitors of advanced glycationprecise role of these factors in the pathogenesis of diabetichave been developed which are at least five times more potentkidney disease is unclear. In the past several years, work fromthan aminoguanidine in vitro [10]. Of particular interest is thea variety of laboratories have clearly demonstrated anthiazolium compound, phenacylthiazolium bromide (PTB), aimportant role for the profibrotic cytokine, transformingnew compound which reacts with and cleaves covalent, AGE-growth factor-b in the pathogenesis of diabetic kidney dis-derived protein cross-links [11]. Such an agent may be par-ease. As TGF-b has the unique characteristic of stimulating aticularly useful clinically in the context of established renalwide variety of matrix molecules that are also upregulated indisease since it may be able to reverse AGE-mediated tissuediabetic nephropathy, it is likely that inhibiting TGF-b in thedamage.early and late stages of diabetic nephropathy would provideclinical benefit.

The evidence supporting a role for TGF-b has been demon-1. Brownlee M. Lilly Lecture 1993. Glycation and diabeticstrated in cell culture, experimental animals and in patientscomplications. Diabetes 1994; 43: 836–841

2. Fu MX, Knecht KJ, Thorpe SR, Baynes JW. Role of oxygen with diabetes [5]. Exposure of mesangial cells and proximalin cross-linking and chemical modification of collagen by glu- tubular cells to high glucose stimulates TGF-b 1 gene expres-cose. Diabetes 1992; 41: 42–48 sion, protein secretion and its bioactivity. Exposure of

3. Brownlee M, Vlassara H, Kooney A, Ulrich P, Cerami A. mesangial cells to high glucose enhances nuclear transcriptionAminoguanidine prevents diabetes-induced arterial wall protein of TGF-b 1 gene, TGF-b 1 promoter activity, and steady-cross-linking. Science 1986; 232: 1629–1632

state TGF-b 1 mRNA levels. The matrix stimulating effects4. Soulis-Liparota T, Cooper M, Papazoglou D, Clarke B, Jerumsof high glucose on types I and IV collagen are blocked byG. Retardation by aminoguanidine of development of albumin-neutralizing antibodies to TGF-b [6 ]. In animal models ofuria, mesangial expansion, and tissue fluorescence in streptozoto-

cin-induced diabetic rat. Diabetes 1991: 40: 1328–1334 diabetes, TGF-b 1 mRNA and protein are elevated in the5. Rumble JR, Cooper ME, Soulis T, Cox A, Wu L, Youssef S, kidneys from rodents with spontaneous diabetes (NOD

Jasik M, Jerums G, Gilbert R. Vascular hypertrophy in experi- mouse and BB rat) [7] as well as in streptozotocin-inducedmental diabetes: role of advanced glycation end products. J Clin diabetes in the mouse [8] and rat [9,10]. In addition, there isInvest 1997; 99: 1016–1027 also upregulation of the TGF-b type II receptor in the6. Makita Z, Vlassara H, Rayfield E, Cartwright K, Friedman E,

STZ-induced diabetic mouse kidney [8].Rodby R, Cerami A, Bucala R. Hemoglobin-AGE: a circulatingNeutralization of TGF-b by antibodies against TGF-b 1, 2marker of advanced glycosylation. Science 1992; 258: 651–653

and 3 inhibits renal hypertrophy, blocks glomerular hyper-7. Abel M, Ritthaler U, Zhang Y, Deng Y, Schmidt AM, Gretentrophy and attenuates diabetes-induced renal expression ofJ, Sernau T, Wahl P, Andrassy K, Ritz E. Expression of

receptors for advanced glycosylated end-products in renal dis- fibronectin and type IV collagen in the early stages of diabetesease. Nephrol Dial Transplant 1995; 10: 1662–1667 [8]. In patients with diabetes, glomerular expression of TGF-

8. Soulis T, Thallas V, Youssef S, Gilbert RE, McWilliam B, b is increased in both early [11] and advanced stages of dia-Murray-McIntosh RP, Cooper ME. Advanced glycation end betic nephropathy [10]. Assaying TGF-b 1 in samples ofproducts and the receptor for advanced glycated end products aortic and renal vein plasma, we recently demonstrated thatco-localise in organs susceptible to diabetic microvascular injury:

non-diabetic patients have renal extraction of TGF-b 1 fromimmunohistochemical studies. Diabetologia 1997; 40: 619–628the circulation, whereas the kidneys of diabetic patients9. Li Y, Mitsuhashi T, Wojciechowicz D, Shimizu N, Li J, Stitt A,uniquely contribute TGF-b 1 to the circulation [12]. AsHe C, Banerjee D, Vlassara H. Molecular identity and distribu-

tion of advanced glycation end products receptors: relationship angiotensin II has been shown to stimulate TGF-b in cultured

Abstracts 1071

of a specific endothelin receptor A antagonist on mRNA levelsmesangial cells and proximal tubular cells, several studiesfor extracellular matrix components and growth factors inhave demonstrated that ACE inhibition reduces renal TGF-bdiabetic glomeruli. Diabetes 1995; 44: 895–8991 expression in animal models of diabetes (reviewed in [4]).

4. Wolf G, Ziyadeh F. The role of angiotensin II in diabeticWe recently found that serum TGF-b1 levels were reducednephropathy: Emphasis on nonhemodynamic mechanisms. Amby captopril treatment in a small sample of patients who were J Kidney Dis 1997; 29: 1: 153–163

enrolled in the Collaborative study [13], and the reduction of 5. Sharma K, Ziyadeh FN. Hyperglycemia and diabetic kidneyserum levels correlated with delayed progression of renal disease: the case for transforming growth factor-b as a keyfunction (JASN (in press)). Apart from the effect of TGF-b mediator. Diabetes 1995; 44: 1139–1146to stimulate matrix proteins, it may also be involved in the 6. Ziyadeh FN, Sharma K, Ericksen M, Wolf G. Stimulation of

collagen gene expression and protein synthesis in murine mesang-altered vasoreactivity of diabetic renal cells. Pretreatmential cells by high glucose is mediated by activation of transformingwith TGF-b of vascular smooth muscle cells [14] and mesang-growth factor-b. J Clin Invest 1994; 93: 536–542ial cells [15] attenuates AII-induced and PDGF-induced

7. Sharma K, Ziyadeh FN. Renal hypertrophy is associated withintracellular calcium mobilization respectively. The effect ofup regulation of TGF-b1 gene expression in diabetic BB rat andTGF-b to inhibit intracellular calcium mobilization may beNOD mouse. Am J Physiol 1994; 267: F1094–F1101

due to enhanced phosphorylation and decreased expression 8. Sharma K, Guo J, Jin Y, Ziyadeh FN. Neutralization of TGF-of the IP3-gated calcium channel, the type I IP3 receptor, in b by anti-TGF-b antibody attenuates kidney hypertrophy andmesangial cells [16 ]. Interestingly, expression of the type I the enhanced extracellular matrix gene expression in STZ-IP3 receptor is reduced in the kidneys of diabetic rats and induced diabetic mice. Diabetes 1996; 45: 522–530

9. Nakamura T, Fukui M, Ebihara I, Osada S, Nagaoka I, Tominodiabetic mice, and by immunostaining there appears to beY, Koide H. mRNA expression of growth factors in glomerulireduced expression of the type I IP3R in glomerular cells andfrom diabetic rats. Diabetes 1993; 42: 450–456vascular smooth muscle cells (JASN (in press)).

10. Yamamoto T, Nakamura T, Noble NA, Ruoslahti E, BorderThus strategies to target and inhibit renal TGF- produc-WA. Expression of transforming growth factor beta is elevatedtion or effect may be beneficial in retarding both the accumu-in human and experimental diabetic nephropathy. Proc Natllation of mesangial matrix as well as vascular dysfunction in Acad Sci USA 1993; 90: 1814–1818

progressive diabetic nephropathy. 11. Iwano M, Atsushi K, Nishino T, Sato H, Nishioka H, Akai Y,In order to propose interventions to block TGF-b an Kurio ka H, Fuji Y, Kanauchi M, Shiiki H, Dohi K.

understanding of the basic biology of the TGF- system is Quantification of glomerular TGF-b 1 mRNA in patients withnecessary. The TGF-b system consists of three major iso- diabetes mellitus. Kidney Int 1996; 49: 1120–1126

12. Sharma K, Ziyadeh FN, Alzahabi B, McGowan TA, Kapoorforms of TGF-b and three receptors [17]. The TGF-b1 iso-S, Kurnik BRC, Kurnik PB, Weisberg LS. Increased renalform is produced by almost all cell types and is unique in thatproduction of transforming growth factor-b 1 in patients withit also circulates in the blood and may have endocrine typetype II diabetes. Diabetes 1997; 46: 854–859effects. It exists in a latent form bound to the TGF-b latency

13. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect ofassociated peptide as well as a2-microglobulin. Binding ofangiotensin-converting-enzyme inhibition on diabetic nephro-

TGF-b by decorin [18], the latency-associated peptide [19] pathy. N Engl Med 1993; 329: 1456–1462and the soluble TGF-b type III receptor [17] may prevent its 14. Zhu Z, Tepel M, Neusser M, Zidek W. Transforming growthbinding to its effector receptors and decrease TGF-b activity. factor beta 1 modulates angiotensin II-induced calcium influx inThe TGF-b type II receptor primarily acts as the ligand bind- vascular smooth muscle. Eur J Clin Invest 1995; 25: 317–321ing receptor at the cell membrane and the complex is then 15. Baffy G, Sharma K, Shi W, Ziyadeh FN, Williamson JR.

Growth arrest of a murine mesangial cell line by transformingrecognized by the signal transducing TGF-b type I receptorgrowth factor b1 is associated with inhibition of mitogen-induced[17].Ca2+ mobilization. Biochem Biophys Res Commun 1995; 210:Overexpression of a dominant negative TGF-b type II378–383receptor has the effect of stimulating pancreatic tumour

16. Sharma K, Wang L, Zhu Y, Bokkala S, Joseph S. Transforminggrowth in overexpressing cells [20]. Activation of the type I growth factor-b1 inhibits type I inositol 1,4,5-trisphosphatereceptor by cross-phosphorylation enhances serine phos- receptor expression and enhances its phosphorylation in mesang-phorylation of the intermediary Smad family of molecules ial cells. J Biol Chem 1997; 272: 14617–14623[21]. Interaction of phosphorylated Smad 2 or 3 with Smad 4 17. Massague J. TGF-b signalling: receptors, transducers, and Madpromotes nuclear translocation and actives gene regulation. proteins. Cell 1996; 85: 947–950

18. Border WA, Noble NA, Yamamoto T, Harper JR, YamaguchiRecently other Smad isoforms, Smad 6 and 7, have beenY, Pierschbacher MD, Rouslahti E. Natural inhibitor of trans-found to inhibit TGF- signalling [22].forming growth factor-b protects against scarring in experi-In conclusion, the evidence implicating TGF-b as anmental kidney disease. Nature 1992; 360: 361–364important factor in the early and late stages of diabetic renal

19. Bottinger E, Factor V, Tsang M, Weatherbee J, Kopp J, Qianinvolvement is very strong. Inhibiting TGF-b systemicallyS, Wakefield L, Roberts A, Thorgeirsson S, Sporn M. Themay lead to untoward effects of increasing cellular prolifera- recombinant proregion of transforming growth factor beta 1

tion and autoimmunity [23]. Greater insight into the signal- ( latency associated peptide) inhibits active transformingling pathways mediating TGF-b induced matrix production +growth factor beta 1 in transgenic mice. Proc Natl Acad Scimay allow for selective blockade of these pathways in target USA 1996; 93: 5877–5882tissues thus limiting the potential for generalized undesired 20. Bottinger E, Jakubczak J, Roberts I, Mumy M, Hemmati P,

Bagnall K, Merlino G, Wakefield L, Source. Expression of aeffects.dominant negative mutant TGF-b type II receptor in transgenicmice reveals essential roles for TGF-b in regulation of growth1. Esposito C, Liu Z, Striker G, Phillips C, Chen N, Chen W,and differentiation in the exocrine pancreas. EMBO J 1997; 16:Kopchick J, Striker L. Inhibition of diabetic nephropathy by a2621–2633GH antagonist: a molecular analysis. Kidney Int 1996; 50:

21. Macias-Silva M, Abdollah S, Hoodless PA, Pirone R, Attisano506–514L, Wrana JL. MADR2 is a substrate of the TGFb receptor and2. Nakamura T, Ebihara I, Fukui M, Osada S, Nagaoka I,its phosphorylation is required for nuclear accumulation andHorikoshi S, Tomino Y, Koide H. Messenger RNA expressionsignalling. Cell 1996; 87: 1215–1224for growth factors in glomeruli from focal glomerular sclerosis.

22. Topper J, Cai J, Qiu Y, Anderson K, Xu Y-Y, Deeds J, FeeleyClin Immunol Immunopathol 1993; 66: 1: 33–423. Nakamura T, Ebihara I, Fukui M, Tomino Y, Koide H. Effect R, Gimeno C, Woolf E, Tayber O, Mays G, Sampson B, Schoen

Abstracts1072

F, Gimbrone M, Falb D. Vascular MADs: two novel cultured mesangial cells inhibited activated PKC activities byMAD-related genes selectively inducible by flow in human high glucose without lowering DAG levels. When LY333531vascular endothelium. Proc Natl Acad Sci USA 1997; 94: was given orally in diabetic rats specifically inhibited the9314–9319 activation of PKC beta1 isoform without decreasing PKC

23. Kulkarni AB, Huh C-G, Becker D, Geiser A, Lyght M, Flanders alpha isoform activation. Glucose-induced increases in arach-KC, Roberts AB, Sporn MB, Ward JM, Karlsson S.idonic acid release, prostaglandin E2 production, and inhibi-Transforming growth factor 1 null mutation in mice causestion of Na+-K+ ATPase activities in the cultured mesangialexcessive inflammatory response and early death. Proc Natlcells were completely prevented by the addition of LY333531.Acad Sci, USA 1993; 90: 770–774Oral feeding of LY333531 prevented the increased mRNAexpression of TGF-b1 and extracellular matrix componentssuch as fibronectin and a1(IV ) collagen in the glomeruli ofdiabetic rats in parallel with inhibition of glomerular PKC

Isozyme-specific inhibition of PKCb. LY333531 new activity. These results suggest that the activation of PKC,theraputic approaches to human disease by modulation of predominately the b isoform by hyperglycaemia in the mes-protein kinase C (PKC ) isozymes angial cells and glomeruli can partly contribute to early renal

dysfunctions by alteration of prostaglandin production andAndreas Pfuetzner; George King, D. Kirk Ways, Michael R.Na+-K+ ATPase activity as well as the chronic pathologicalJirousek; Lilly Research Laboratories, Eli Lilly & Co. andchanges by the overexpression of TGF-b1 and extracellularJoslin Diabetes Center, Harvard Medical Schoolmatrix components genes.

Protein kinase C (PKC) is a family of closely related serine These results taken in context suggest selective inhibitionand threonine kinases. Overactivation of some PKC isozymes of the PKCb isozymes by LY333531 may offer a newhas been postulated to occur in several diseases states includ- approach to address kidney disease as well as the microvascu-ing diabetic complications. Hyperglycaemia leads to over- lar complications and cardiomyopathy of diabetes [5].activation of PKC and selectively to hyperactivation of PKCbthrough elevated diacylglycerol levels. The discovery and 1. Jirousek MR, Gillig JR, Gonzalez CM, Heath WF, McDonalddevelopment of LY333531 offers an oppertunity to test these JH 3rd, Neel DA, Rito CJ, Singh U, Stramm LE, Melikian-hypothesis. LY333531 selectively inhibits the PKCbI (IC50= Badalian A, Baevsky M, Ballas LM, Hall SE, Winneroski LL,4.7 nM) and PKCbII (IC50=5.9 nM) isozymes with about Faul MM. (S)–13-[(dimethylamino)methyl ]–10,11,14,15-tetra-

hydro–4,9:16, 21-dimetheno–1H, 13-H-dibenzo [e,k]70-fold selectivity for the beta isoforms relative to PKCa [1].pyrrolo[3,4-h][1,4,13]oxa diazacyclohexadecene–1,3[2H)-d ioneThe inhibitor exhibits ATP-dependent competitive inhibition(LY333531) and related analogues: isozyme selective inhibitorsof PKCbI and is selective for PKC in comparison to otherof protein kinase C beta. J Med Chem 1996; 39: 2664–2671ATP dependent kinases (protein kinase A, calcium calmodu-

2. Ishii H, Jirousek MR, Koya D, Takagi C, Xia P, Clermont A,lin, caesin kinase, src tyrosine kinase). Bursell SE, Kern TS, Ballas LM, Heath WF, Stramm LE, FeenerThe in vitro selective inhibition of PKCb by LY333531 has EP, King GL. Amelioration of vascular dysfunctions in diabetic

also been demonstrated at the cellular level [2]. When admin- rats by an oral PKC beta inhibitor. Science 1996: 272: 728–731istered orally, LY333531 ameliorated the glomerular filtration 3. Xia P, Aiello LP, Ishii H, Jiang ZY, Park DJ, Robinson GS,

Takagi H, Newsome WP, Jirousek MR, King GL. Character-rate, albumin excretion rate, and retinal circulation in dia-ization of vascular endothelial growth factor’s effect on thebetic rats in a dose-responsive manner, in parallel with itsactivation of protein kinase C, its isoforms, and endothelial cellinhibition of PKC activities. The observation that PKCbgrowth. J Clin Invest 1996; 98: 2018–2016regulates vascular endothelial growth factor (VEGF )-

4. Koya D, Jirousek MR, Lin YW, Ishii H, Kuboki K, King GL.induced mitogenesis suggests that a selective PKCb inhibitorCharacterization of protein kinase C beta isoform activation on

such as LY333531 may be useful in VEGF-mediated disease the gene expression of transforming growth factor-beta, extracel-states [3]. VEGF induced concentration-and time-dependent lular matrix components, and prostanoids in the glomeruli ofincreases in PKC activation. VEGF stimulation increased the diabetic rats. J Clin Invest 1997; 100: 115–126content of Ca(2+)-sensitive PKC isozymes (a and bII ) in the 5. Wakasaki W, Koya D, Schoen FJ, Jirousek MR, Ways DK, Hoit

BD, Walsh RA, King GL. Targeted overexpression of proteinmembrane fractions, while no changes were observed forkinase Cb 2 isoform in myocardium causes cardiomyopathy. ProcPKC isoforms delta and epsilon.Natl Acad Sci USA 1997: 94: 9320–9325The stimulation of PKC activity by VEGF was preceded

by the activation of phospholipase C gamma (PLCc). VEGFincreased phosphatidylinositol 3-kinase activity 2.1-fold,which was inhibited by wortmannin, a phosphatidylinositol3-kinase inhibitor, without decreasing the VEGF-induced Progression in inherited renal diseaseincrease in PKC activity or endothelial cell growth. VEGF’s

Jean Pierre Grunfeld, E. Plaisier, P. Jungers; Service demitogenic effect was inhibited by LY333531, in a concentra-Nephrologie, Hopital Necker, 75015 Paris, Francetion-dependent manner. In contrast, antisense PKCa oligonu-

cleotides enhanced VEGF-stimulated cell growth with a The potential mechanisms of renal progression are diverse insimultaneous decrease of 70% in PKCa protein content. Thus the various inherited renal diseases: e.g. cyst progression andVEGF appears to mediate its mitogenic effects partly through compression of adjacent parenchyma in autosomal dominantthe activation of the PLCc and PKC pathway, involving pre- polycystic kidney disease (ADPKD); persistence and exten-dominately PKCb isoform activation in endothelial cells. sion of a1/a2 chains of type IV collagen in GBM, in the place

A spectrum of biochemical and molecular abnormalities of a3/a4/a5 (IV ); focal segmental glomerulosclerosis in type Iassociated with chronic changes induced by glucose or dia- glycogen storage disease; glycolipid deposition, most particu-betes in the cultured mesangial cells and renal glomeruli that larly in vessels, leading predominantly to ischaemic nephro-can be prevented by LY333531 [4]. Hyperglycaemia increased pathy in Fabry’s disease. There is no single mechanism ofDAG level in cultured mesangial cells exposed to high con- progression in inherited disorders.centrations of glucose and activated PKCa and b1 isozymes In ADPKD, two main determinants of progression have

been identified: (1) PKD1 vs PKD2 disease, the renal progres-in the renal glomeruli of diabetic rats. LY333531 added to

Abstracts 1073

sion being slower in the PKD2 type; interactions between The hypotheses which have formed the basis for studies ofPKD1 and PKD2 genes have been recognized, and these may renal pathophysiology have rarely been ‘competing’. Rather,be crucial to modulate the rate of progression. (2) The sex of they have invariably emerged as being complementary whichthe patient, the renal progression being slower in women. is not surprising in that the processes which cause disease areAmong 126 ADPKD patients in ESRD at Necker, from 1989 almost always multifactorial in nature. Renal scarring is noto 1996, 42% were females; in the 26 patients in ESRD at 65 exception to this.years or more, 65% were females, and all 5 in ESRD at 70 The central issue in renal scarring, which is usually second-years or more were female. Overall, women (n=53) were in ary to glomerular disease, is: Why does an acute insult notESRD at 56+11 years and men (n=73) at 53.9+9 years. resolve but rather translate itself into a chronic destructiveLow-protein diet, strict control of blood pressure, and ACE process? To answer this question it is useful to distinguishinhibition have been shown to have no significant effect on between (1) factors that per se are responsible for chronicityrenal progression. However, the effects of earlier intervention i.e. the ‘causes’ of the scarring process, (2) factors that affectin the course of the disease have not been tested. the rate of scarring, and (3) factors which mediate the

Alport syndrome is clinically and genetically heterogen- scarring process.eous. In X-linked disease (due to mutations of the gene encod- Hypotheses that address the ‘cause of chronicity’ musting for a5 chain of type IV collagen), the rate of progression include the following:in males contrasts in juvenile-type (EDRD <30 years) andadult-type (ESRD >30 years) families. 1. The initial injurious agent persists and causes chronic

The rate of decline of creatinine clearance is also slower in damage. This could apply to diabetes but probably cannotthe latter kindreds. No clear-cut relationship has been found be invoked in most primary glomerulopathies.so far between phenotype and genotype. The rate of progres- 2. Chronic glomerular scarring is the result of haemo-sion is rapid in autosomal recessive disease (defects of a3 or dynamic adaptations that are set in train by the initiala4 (IV )). So far it has not been tested whether this rate is insult and which, if sufficiently severe, will destroy suffi-similar or not in female and male siblings from the same cient filtration surface area to cause the adaptation to befamilies. In anecdotal cases of Alport syndrome, ACE inhibi- persistent in surviving glomeruli. The ‘adaptation’ thustion has been claimed to be renoprotective, which would not becomes the injurious agent.be surprising in a primary glomerular disease. No prospective 3. Tubulointerstitial injury follows the initial glomerularstudy is available. insult and it is this component that is self-perpetuating.The mechanisms of progression are unknown. Kalluri For this hypothesis for be valid, there must be a linket al. have suggested that isoform switching (from a1/a2 to

between the glomerular injury (or adaptation) and thea3/a4 (IV )) is developmentally arrested in X-linked Alporttubulointerstitial compartment, and injury to the latterdisease. This defect would lead to increased susceptibility ofmust be irreversible and progressive.GBM to endoproteolysis. Glomerular accumulation of a1/a2

What causes tubulointerstitial injury associated with glom-(IV ) by itself could be deleterious. The molecular mechanismerular disease? The two principal ideas are that:of progression might be of great interest not only for Alport(1) Filtration of proteins, lipids and other molecules injuresyndrome, but also for other glomerular diseases.

the tubules and this sets up the scarring process. This isSimilarly, the mechanism of progression of interstitial fib-supported by the recognition that the magnitude of pro-rosis in juvenile nephronophthisis, an autosomal recessiveteinuria is a major predictor of the rate of progressiondisease, may provide interesting insight into more generaland reversal of proteinuria slows progression.mechanisms of initiation and progression of interstitial

(2) The ‘chronic hypoxia hypothesis’, which posits that thefibrosis.same alterations in flow and pressure that affect the glom-

Klahr S, Breyer JA, Beck GJ et al. Dietary protein restriction, blood erulus (albeit of lesser magnitude) also cause microvascu-pressure control, and the progression of polycystic kidney disease. lar disease in the interstitium. This limits oxygen deliveryJ Am Soc Nephrol 1995; 5: 2037–2047 to tubular cells, which respond by producing cytokinesChoukroun G, Itakura Y, Albouze G et al. Factors influencing

and chemokines which initiate and perpetuate fibrosis.progression of renal failure in autosomal dominant polycysticAggravation of this injury occurs in hypertension andkidney disease. J Am Soc Nephrol 1995; 6: 1634–1642where there is a primary vascular component to theQian F, Germino FJ, Cai Y et al. PKD1 interacts with PKD2

through a probable coiled-coil domain. Nat Genet 1997; 16: disease (amyloid, diabetes) and control of blood pressure179–183 limits interstitial injury.

Tsiokas L, Kim E, Arnould T et al. Homo- and hetrodmeric Why does interstitial scarring not resolve?interactions between the gene products of PKD1 and PKD2. Proc The fact that some glomerular diseases progress even whenNatl Acad Sci USA 1997; 94: 6965–6970

proteinuria is completely reversed or when blood pressure isKashtan C, Michael AF. Alport syndrome. Kidney Int 1996; 50:well controlled must mean that these are not ‘causes’ of the1145–1163progression but rather modulators of its rate.Cohen EP, Lemann J Jr. In hereditary nephritis angiotensin-

converting enzyme inhibition decreases proteinuria and may slow The chronic hypoxia hypothesis critically requires thatthe rate of progression. Am J Kidney Dis 1996; 27: 199–203 interstitial microvessels be surrounded by cells which can

Kalluri R, Shield CFI, Todd P et al. Isoform switching of type IV sense and respond to hypoxia once the scarring process hascollagen is developmentally arrested in X-linked Alport syndrome been initiated. We have shown that human proximal tubularleading to increased susceptibility of renal basement membranes

cells in culture produce cytokines in an oxygen-sensitiveto endoproteolysis. J Clin Invest 1997; 99: 2470–2478manner (endothelin I, VEGF, TGFb). Others have shownthat such cells also release chemokines that attract mono-nuclear cells which contribute to inflammation and scarring.

Competing hypothesis in renal scarring: further support for the We have further shown that hypoxia is directly fibrogenic tochronic hypoxia hypothesis proximal tubular cells and renal interstitial fibroblasts which

produce collagen, decrease their matrix metalloproteinase,Leon G. Fine, Chrystalla Orphanides, Jill T. Norman; Jointand increase their TIMP activities.Department of Medicine University College London and

Royal Free Hospital Schools of Medicine, London, UK Because the fibrosing process per se obliterates

Abstracts1074

microvessels, i.e. the ‘healing of a scar’, the hypoxia is per- dihydropyridine calcium-channel blockers (CCBs), verapamiland diltiazem, preserve renal morphology in a manner similarpetuated and the parenchymal cells which are adjacent toto ACE inhibitors. This subclass of CCBs affects the synthesissuch ischaemic areas respond by further collagen depositionof various matrix proteins such as heparan sulphate and glu-and cytokine release. This in turn causes further microvascu-coaminoglycan as well as fibroblast turnover [7]. Preventionlar loss and hypoxia and the process is self-perpetuating.of mesangial expansion or glomerulosclerosis has not beenThe chronic hypoxia hypothesis therefore requires that ashown for either blockers or most dihydropyridine CCBscritical degree of initial hypoxic damage to the tubulointerstit-[3,6 ].ium is sufficient to set the vicious cycle in process. If the initial

Dihydropyridine CCBs do not protect against morpholo-injury is not sufficiently severe, small or healing scars willgical progression of renal disease. Studies using 24-h bloodresult and the disease will not be progressive.pressure monitoring in a rat model of renal insufficiency ran-Studies are in progress which will establish how lowereddomized animals to either a dihydropyridine, amlodipine, anoxygen tension triggers collagen production independently ofACE inhibitor, benazepril or a combination of each in a lowermediators such as TGF-b . The relevant molecules would bedose to achieve a similar level of blood pressure reduction.logical targets for interventional therapy.Amlodipine did not reduce development of glomerulo-sclerosis or proteinuria, while the ACE inhibitor and thecombination had similar benefits on these parameters [8].

This lack of protection by dihydropyridine CCBs is alsoCombination therapy for hypertension and renal disease in seen clinically and relates to a lack of effect on glomerulardiabetes membrane permeability, in spite of blood-pressure reduction

[9]. While, many factors may contribute to the disparateGeorge L. Bakris; Rush University Hypertension Center,effects of the different subclasses of CCBs on both surrogateRush-Presbyterian-St. Luke’s Medical Center, Chicago, ILend-points as well as progression of renal disease, it is clear60612, USAthat arterial pressure reduction is the ultimate goal that uni-

More than 75% of individuals with diabetes, microalbuminu- formly leads to preservation of renal function. Thus, since theria and hypertension require two or more antihypertensive majority of patients with diabetic nephropathy require moremedications to attain the newly recommended level of arterial than one medication to reduce arterial pressure, a combina-pressure control, i.e.<130/85 mmHg [1]. To achieve this tion of antihypertensive agents, individually shown to reduceblood-pressure goal, it is common to add a medication whose arterial pressure and proteinuria as well as preserve renalantihypertensive action potentiates the initially selected drug morphology and function would be preferred. Since prelimin-[2]. To further improve compliance and reduce drug side ary evidence supports the notion that both ACE inhibitorseffect profiles, fixed-dose combinations of antihypertensive and non-dihydropyridine CCBs slow progression of diabeticdrugs with complementary modes of action have been renal disease to a greater extent than conventional bloodrecently developed [2]. These medications combine a lower pressure lowering agents, it is predicted that a combination ofdose of two different antihypertensive drugs that, in a fixed these two classes of agents will provide better renal protectiondose combination, reduce arterial pressure to a greater extent than either alone. Unfortunately, few animal or human stud-than either alone. ies have examined this hypothesis.

While blood-pressure reduction is critical for the preserva- Of the few animal studies that have evaluated the effects oftion of renal function, it is also important to select antihyper- an ACE inhibitor/CCB combination on various aspects oftensive agents that have unique effects on the kidney. Early renal disease, only one has actually controlled for blood-changes observed in the diabetic kidney include an increase in pressure differences [9]. Thus a meaningful comparisonefferent arteriolar tone and a loss of autoregulation. These cannot be made between the combination and its individualchanges lead to an increase in intraglomerular capillary pres- components. As previously mentioned this study demon-sure, which results in cell stretch and activation of various strated that combination therapy with a fixed-dose of an ACEautocrine and paracrine factors associated with tissue inhibitor, benazepril, and dihydropyridine CCB, amlodipine,injury [3]. In addition, glomerular membrane permeability is prevented progression to glomerulosclerosis seen with amlod-increased and microalbuminuria ensues, a surrogate marker ipine alone. It also reduced proteinuria to a greater extentfor the presence and progression of diabetic nephropathy [3]. than amlodipine alone. However, the effects on proteinuriaThe earliest morphologic change is mesangial matrix expan- while greater than benazepril alone were not statisticallysion and in some cases interstitial inflammation, the latter different.portends a poor renal prognosis [3,4]. In contrast to these findings, two animal studies that com-

Experimental evidence demonstrates that reductions in bined the effects of a non-dihydropyridine CCB with an ACEintraglomerular capillary pressure, through either profound inhibitor demonstrated a potentiation of the ACE inhibitorreductions in arterial pressure or dilation of the efferent art- associated reduction in proteinuria as well as relativelyeriole, slow progression of diabetic renal disease [3,5]. ACE greater preservation of renal morphology [5,10]. These effectsinhibitors reduce efferent arteriolar tone and intraglomerular were noted at similar levels of blood-pressure reduction.pressure. In clinical studies, ACE inhibitors demonstrate a Moreover, one of the two studies showed greater morpholo-consistent and persistent reduction in proteinuria as well as gical protection of the glomerulus, in the absence of blood-an attenuated progression of both type 1 and type 2 diabetes pressure control, using the fixed-dose combination of verapa-associated nephropathy [3,6 ]. Moreover, while blood pres- mil with trandolopril in comparison to their individualsure reduction itself slows progression of early diabetic components [10]. The potentiating antiproteinuric effects ofnephropathy, the ACE inhibitors appear to preserve renal a fixed-dose combination of verapamil and trandolopril, vs itsfunction to an even greater extent than other agents [5,6 ] do. individual components, has also been observed in a recent

Antihypertensive agents that do not reduce intraglomerular randomized clinical study of patients with type 2 diabetespressure may still slow progression of diabetic nephropathy. associated nephropathy and comparable blood-pressure con-This may relate to their specific effects on synthesis or trol [11].

Additionally, a 4-year follow-up study of patients with typedegradation of mesangial matrix proteins. Notably, non-

Abstracts 1075

2 diabetes associated nephropathy, randomized to either ver- Rational treatment in diabetic renal disease, what is theevidence?apamil or lisinopril, alone or each in a reduced dose combina-

tion, demonstrated that combination therapy had a slower Hans-Henrik Parving; Steno Diabetes Center, Gentofte,rate of decline in renal function, a lower amount of pro- Denmarkteinuria and the lowest side-effect profile [12]. No other long

Diabetic nephropathy is a clinical syndrome characterized byterm clinical studies have formally evaluated either fixed-dosepersistent albuminuria (>300 mg/24 h), a relentless decline inor reduced dose combination therapy on progression of dia-glomerular filtration rate (GFR), and elevated systemic bloodbetic renal disease.pressure. The prevalence of abnormally elevated albuminIt is critically important to remember that effective arterialexcretion rate (>30 mg/24 h) is approximately 40% inpressure reduction needs to be achieved in diabetic patientspatients with both insulin-dependent diabetes mellitus (typeto not only preserve renal function but also reduce cardiovas-1) and non-insulin-dependent diabetes mellitus (type 2).cular mortality, the latter being the primary cause of death inDiabetes has become the leading cause of end-stage renalthese patients. Evidence supports the concept of arterial pres-failure in the United States, Japan, and in Europe.sure reduction to levels of less than 130/85 mmHg maximallyIdentification of patients at high risk of developing diabeticprotects the kidney and reduces cardiovascular mortality.nephropathy is possible by screening for microalbuminuriaDiabetic patients have difficulty in controlling blood pressure;(30 to 300 mg/24 h) [1,2]. Randomized controlled trials inthus, combination therapy would be highly recommended. Ofnormotensive type 1 and type 2 diabetic patients with persist-the many possible combinations, a non-dihydropyridineent microalbuminuria indicate that angiotensin-convertingCCB/ACE inhibitor combination may be preferred not onlyenzyme (ACE) inhibitors diminish urinary albumin excretionbecause of a low side-effect profile but also the fact that eachrate and postpone and may even prevent progression to clin-agent has been individually shown to reduce cardiovascularical overt diabetic nephropathy [3–5]. These findings suggestmortality following myocardial infarctions [3].that screening and intervention programmes are likely to haveIn the light of these few studies, however, there are insuffi-life-saving effects and lead to considerable economic savingscient data to assess whether combination antihypertensive[6 ].therapy offers a distinct advantage over its individual com-

Previous studies have demonstrated that the diabeticponents to further slow progression of diabetic nephropathypatients with the most marked proteinuria have the worseat a given level of blood-pressure reduction.prognosis. Recently it has been suggested that proteinuria isnot simply a marker of the extent of glomerular damage, but1. National High Blood Pressure Education Program Workingproteinuria per se may contribute to glomerular damage.Group on Hypertension and Renal Disease. 1995 Update of theFurthermore, a decrease in albuminuria shortly after onset ofworking group reports on chronic renal failure and renovascularACE-inhibition predicts an attenuated rate of fall in GFR inhypertension. Arch Intern Med 1996; 156: 1938–1947

2. Epstein M, Bakris GL. Newer approaches to antihypertensive diabetic nephropathy. The importance of this finding is high-therapy using fixed dose combination therapy: future perspect- lighted by the demonstration that ACE-inhibition has anives. Arch Intern Med 1996; 156: 1969–1978 antiproteinuric effect independent of the effect on systemic

3. Bakris GL, Mehler P, Schrier R. Hypertension and diabetes. In: blood pressure.Schrier RW, Gottschalk CW (eds.) Diseases of the Kidney 6th Impaired nocturnal decline in blood pressure is more pre-edn. Little Brown, 1996; 1281–1328

valent in patients with diabetic nephropathy and autonomic4. Steffes MW, Osterby R, Chavers B, and Mauer SM. Mesangialneuropathy. This may contribute to the enhanced cardiovas-expansion as a central mechanism for loss of kidney function incular morbidity found in that condition. Raised blooddiabetic patients. Diabetes 1989; 38: 1077–1081pressure accelerates both the development and the progres-5. Gaber L, Walton C, Brown S, Bakris, GL: Effects of different

antihypertensive treatments on morphologic progression of dia- sion of nephropathy in both type 1 and type 2 diabeticbetic nephropathy in uninephrectomized dogs. Kidney Int 1994; patients [7]. Arterial blood pressure thus seems to have a46: 161–169 complex relationship with diabetic nephropathy—nephro-

6. Maki DD, Ma JZ, Louis TA, Kasiske BL: Effect of antihyper- pathy raising blood pressure and blood pressure acceleratingtensive agents on the kidney. Arch Intern Med 1995; 155: the course of nephropathy. Effective blood pressure reduction1073–1082

with b-blockers and/or angiotensin-converting enzyme inhib-7. Jyothirmayi GN, Reddi AS. Effect of diltiazem on glomerularitors combined with diuretics reduces albuminuria, delays theheparan sulfate and albuminuria in diabetic rats. Hypertensionprogression of nephropathy and postpones renal insufficiency1993; 21: 765–802

8. Bakris GL Griffin K, Picken M, Bidani A. Combined effects of in diabetic nephropathy [7–9]. Originally Bjorck et al. [8]an angiotensin converting enzyme inhibitor and a calcium demonstrated that treatment with ACE inhibition combinedantagonist on renal injury. J Hypertens (in press) with diuretics can reduce the rate of decline in kidney function

9. Bakris GL. Combination therapy for hypertension and renal in type 1 diabetic patients with moderately advanced diabeticdisease in diabetes. In: Mogensen CE (ed.) The Kidney and nephropathy (mean GFR 47 ml/min/1.73 mO) more thanHypertension in Diabetes Mellitus, 3rd edn. Kluwer, Boston,

equally effective antihypertensive treatment with metoprolol1997; 561–568and diuretics. This finding suggests renoprotection—a bene-10. Munter K, Hergenroder S, Jochims K, Kirchengast M.ficial effect on kidney function (and structure) above andIndividual and combined effects of verapamil or trandolapril onbeyond that expected from the blood-pressure-reducing effectglomerular morphology and function in the stroke prone rat. J

Am Soc Nephrol 1996; 7: 681–686 alone. This suggestion has been confirmed and extended in a11. Bakris GL, Weir, MR, DeQuattro V, Rosendorff C, MacMahon recent randomized double blind study comparing the effects

G. Renal hemodynamic and antiproteinuric response to an ACE of captopril versus placebo (receiving conventional antihyper-inhibitor, trandolopril or calcium antagonist, verapamil, alone tensive treatment) in type 1 diabetic patients with diabeticor in fixed dose combination in patients with diabetic nephro- nephropathy, whose baseline serum creatinine concentrationpathy: a randomized multicentered study. Abstract. J Am Soc

was above 133 mmol/l [9]. The introduction of effective anti-Nephrol 1996; 7: 1546hypertensive treatment has increased the median survival time12. Lash JP, Bakris GL. Effects of ACE inhibitors and calciumto more than 16 years as compared to 5 and 7 years inantagonists alone or combined on progression of diabetic

nephropathy. Nephrol Dial Transplant 1995; 10: [Suppl. 9]: 56–62 untreated patients in the past [10]. The potential beneficial

Abstracts1076

effect of antihypertensive drugs, e.g. ACE inhibitors, on pro- TGF-b system and growth in LLC-PK1 cells, a porcinekidney cell line analogous to the proximal tubule cell [70]. Ingression of kidney function in normotensive type 1 and type

2 diabetic patients suffering from diabetic nephropathy has this cell system high-glucose increased TGF-b 1 mRNA,TGF-b type I and II receptor protein expression and cellularnot been settled.hypertrophy, while cellular mitogenesis was inhibited.Captopril dose-dependently decreased TGF-b type I and II1. Parving H-H, Oxenbøll B, Svendsen PA, Christiansen JS,receptor protein expression and cellular hypertrophy,Andersen AR. Early detection of patients at risk of developingincreased cellular hyperplasia, while TGF-b 1 mRNA wasdiabetic nephropathy. Acta Endocrinol (Copenh) 1982; 100:

550–555 unchanged [70]. So far no published data have examined the2. Viberti GC, Hill RD, Jarrett RJ, Argyropoulos A, Mahmud U, possible effect of ACE inhibition on the intrarenal changes in

Keen H. Microalbuminuria as a predictor of clinical nephro- the various TGF-b isoforms (TGF-b 1,2,3] and TGF-b typepathy in insulin-dependent diabetes mellitus. Lancet 1982; 1: I, type II, type III receptors (TGF-b RI, TGF-b RII, TGF-b1430–1432 RIII ) in experimental diabetes in vivo. Accordingly, we per-3. Mathiesen ER, Hommel E, Giese J, Parving H-H. Efficacy of

formed a study in which immunocytohistochemistry was per-captopril in postponing nephropathy in normotensive diabeticformed on kidney sections from non-diabetic andpatients with microalbuminuria. Br Med J 1991; 303: 81–87streptozotozin (STZ)-diabetic rats after 2 and 4 weeks treat-4. Viberti GC, Mogensen CE, Groop L, Pauls JF, the Europeanment with an ACE-inhibitor (enalapril ) or placebo, usingMicroalbuminuria Captopril Study Group. Effect of captopril

on progression to clinical proteinuria in patients with insulin- specific TGF-b 1,2,3 and TGF-b RI, TGF-b RII, TGF-b RIIIdependent diabetes mellitus and microalbuminuria. JAMA 1994; antibodies. Enalapril partially prevented the diabetes associ-271: 275–279 ated renal hypertrophy, while no effect was seen on kidney

5. Ravid M, Savin H, Jutrin I, Bental T, Katz B, Lishner M. weight in non-diabetic animals. In addition, the diabetes asso-Long-term stabilizing effect of angiotensin-converting enzyme ciated increase in 24-h urinary albumin excretion (UAE) wasinhibition on plasma creatinine and on proteinuria in normotens-

fully prevented by enalapril with no significant effects in non-ive type II diabetic patients. Ann Intern Med 1993; 118: 577–581diabetic animals. Enalapril therapy had no effect on body-6. Mogensen CE, Keane WF, Bennett PH, Jerums G, Parvingweight, metabolic control or food consumption in eitherH-H, Passa P et al. Prevention of diabetic renal disease withgroups. Immunoreactivity of the glomerular TGF-b1 isoformspecial reference to microalbuminuria. Lancet 1995; 346:

1080–1084 revealed a transient decrease after 14 days with a normaliza-7. Parving H-H, Rossing P. The use of antihypertensive agents in tion at 30 days in untreated diabetic animals. In enalapril-

prevention and treatment af diabetic nephropathy. Curr Opin treated diabetic animals the immunoreactivity stayed belowNephrol Hypertens 1994; 3: 292–300 control level at all time points.

8. Bjorck S, Johnsen SA, Nyberg G, Aurell M. Contrasting effects The immunoreactivity of the glomerular TGF-b2 isoformof enalapril and metoprolol on proteinuria in diabetic nephro-

increased over 30 days in untreated diabetic animals, whilepathy. Br Med J 1990; 300: 904–907this increase was only partially abolished by enalapril-9. Lewis E, Hunsicker L, Bain R, Rhode R. The effect of angio-treatment. The glomerular TGF-b 3 immunoreactivitytensin-converting-enzyme inhibition on diabetic nephropathy. Nincreased over 30 days in untreated diabetic animals with noEngl J Med 1993; 329: 1456–1462

10. Parving H-H, Jacobsen P, Rossing K, Smidt UM, Hommel E, major effect of enalapril-treatment. The immunoreactivity ofRossing P. Benefits of long-term antihypertensive treatment on glomerular TGF-b RI and TGF-b RIII isoforms wasprognosis in diabetic nephropathy. Kidney Int 1996; 49: increased in untreated diabetic animals after 14 and 30 days1778–1782 when compared with control rats, while enalapril treatment

decreased the immunoreactivity of both receptors to valuesbelow non-diabetic control level at all time points. Similarly,glomerular TGF-b RII immunoreactivity increased over 30days in untreated animals when compared to non-diabeticEffect of ACE inhibition on the altered intrarenal transformingcontrols, while enalapril therapy in diabetic animals wasgrowth factor-b(TGF-b) system in experimental diabetesassociated with a dramatic decrease over 30 days to almost

Charlotte Hill, Ann Logan, Michael C. Sheppard, Henning undetectable levels.Grønbæk, Allan Flyvbjerg; Department of Medicine, In conclusion, our data suggesting that the TGF-b axisUniversity of Birmingham, Birmingham, UK, and Medical operating through a complex intrarenal system, may be aDepartment M, Diabetes & Endocrinology, Aarhus significant mediator of the renal changes observed in experi-Kommunehospital, Aarhus University Hospital, Aarhus, mental diabetes. Moreover, ACE-inhibition has pronouncedDenmark inhibitory effects on the elevated levels of the TGF-b recep-

tors required for intracellular signalling through this growthSeveral growth factors [1], including the transforming growthfactor system. The present findings suggest a possible newfactor-b (TGF-b) system [2–7], have been suggested to playmechanism of action for ACE inhibitors.a role in the development of diabetic nephropathy. In vitro

high glucose concentrations increase TGF-b 1 mRNA levelsin both cultured mesangial cells and proximal tubular cells 1. Flyvbjerg A. Role of growth hormone, insulin-like growth factors[2,3]. In addition in vivo experiments have been published on (IGFs) and IGF-binding proteins in the renal complications of

diabetes. Kidney Int 1997; 52: [Suppl. 60]: S12–S19changes in the endogenous renal TGF-b axis in various2. Ziyadeh FN, Chen Y, Davila A, Goldfarb S. Self limited stimula-animal models of experimental diabetes [4–7].

tion of mesangial cell growth in high glucose: autocrine activationIt has been suggested that activation of the renal TGF-bof TGF-b reduces proliferation but increases mesangial matrix.system in diabetes may be mediated, beside a direct stimula-Kidney Int 1992; 42: 647–656tory effect of hyperglycaemia per se [2,3], through activation

3. Rocco MV, Chen Y, Goldfarb S, Ziyadeh FN. Elevated glucoseof the renin–angiotensin system. Accordingly, exposure of stimulates TGF-b gene expression and bioactivity in proximalmesangial cells in vitro to angiotensin II stimulates the expres- tubules. Kidney Int 1992; 41: 107–114sion of TGF-b and extracellular matrix proteins [8]. 4. Shankland SJ, Scholey JW. Expression of transforming growthFurthermore a recent study examined the effect of an ACE factor 1 during diabetic renal hypertrophy. Kidney Int 1994;

46: 430–442inhibitor (captopril ) on high-glucose induced changes in the

Abstracts 1077

5. Sharma K, Ziyadeh FN. Renal hypertrophy is associated with patients, percentage reduction in proteinuria was inverselyupregulation of TGF-b1 gene expression in diabetic BB rat and correlated with decline in GFR (P=0.035) and predicted theNOD mouse. Am J Physiol 1994; 67: F1094–F1101 reduction in risk of doubling of baseline creatinine or end-

6. Pankewycz OG, Guan JX, Bolton WK, Gomez A, Benedict JF. stage renal failure (18 ramipril vs 40 placebo, P=0.04). TheRenal TGF-b regulation in spontaneously diabetic NOD mice risk of progression was still significantly reduced after adjust-with correlations in mesangial cells. Kidney Int 1994; 46: 748–758

ment for changes in systolic (P=0.04) and diastolic (P=0.04)7. Hill C, Flyvbjerg A, Thomas C, Sheppard MC, Grønbœk H,blood pressure, but not after adjustment for changes in pro-Logan A. Localisation of TGF-b isoforms and their receptorsteinuria. Blood-pressure control and the overall number of(R) in acute and chronic experimental diabetic nephropathy.

Abstract. J Endocrinol 1997; 152: P85 cardiovascular events were similar in the two treatment8. Kagami S, Border WA, Miller DE, Noble NA. Angiotensin II groups.

stimulates extracellular matrix protein synthesis through induc-tion of transforming growth factor- expression in rat glomerularmesangial cells. J Clin Invest 1994; 93: 2431–2437 Interpretation

9. Guh JY, Yang ML, Yang YL, Chang CC, Chuang LY. CaptoprilIn chronic nephropathies with proteinuria of 3 g or more perreverses high-glucose-induced growth effects on LLC-PK1 cells24 h, ramipril safely reduces proteinuria and the rate of GFRpartly by decreasing transforming growth factor- receptor protein

expression. J Am Soc Nephrol 1996; 7: 1207–1215 decline to an extent that seems to exceed the reductionexpected for the degree of blood-pressure lowering.

Ellis A. The natural history of Bright’s disease; clinical, histologicaland experimental observations Ill: The vicious cycle in chronic

Randomized placebo-controlled trial of effect of ramipril on Bright’s disease. Lancet 1942; 1: 72–76decline in glomerular filtration rate and risk of terminal renal Lewis EJ, Hunsicker LG, Baim RP, Rohde RD, for the collaborative

study group. the effect of angiotensin-converting-enzyme inhibi-failure in proteinuric, non-diabetic nephropathytion on diabetic nephropathy. N Engl J Med 1993; 329: 1456–1462Giuseppe Remuzzi; Department of Transplant Immunology Maschio G, Alberti D, Janin G et al. Effect of the angiotensin

and Innovative Antirejection Therapies, Ospedaii Riuniti converting enzyme inhibitor benazepril on the progression ofdi Bergamo, Mario Negri Institute for Pharmacological chronic renal insufficiency. N Engl J Med 1996; 334: 939–945Research, Bergamo, Italy Giatras I, Lau J, Levey AS. Effect of angiotensin-converting enzyme

inhibitors on the progression of non-diabetic renal disease: ameta-analysis of randomized trials. JASN 1996; 7: 1318 (A) andAnn Intern Med (in press)Background

van Essen GG, Rensma PL, de Zeeuw D et al. Association betweenIn diabetic nephropathy, angiotensin-converting enzyme angiotensin-converting-enzyme gene polymorphism and failure of(ACE) inhibitors have a greater effect than other antihyper- renoprotective therapy. Lancet 1996; 347: 94–95tensive drugs on proteinuria and the progressive decline in Parving H-H, Jacobsen P, Tarnow L et al. Effect of deletionglomerular filtration rate (GFR). Whether this difference polymorphism of the angiotensin converting enzyme gene on

progression of diabetic nephropathy during inhibition of angio-applies to progression of non-diabetic proteinuric nephro-tensin converting enzyme: observational follow-up study. Br Medpathies is not clear. The Ramipril Efficacy In NephropathyJ 1996; 313: 591–594study of chronic non-diabetic nephropathies aimed to address

Hannedouche T, Landais P, Goldfarb B et al. Randomised controlledwhether glomerular protein traffic influences renal-diseasetrial of enalapril and beta-blockers in non-diabetic chronic renalprogression, and whether an ACE inhibitor was superior to failure. Br Med J 1994; 309: 833–836

conventional treatment, with the same blood-pressure con- van Essen GG, Apperloo AJ, Rensma PL, Sluiter WJ, de Zeeuw D,trol, in reducing proteinuria, limiting GFR decline, and de Jong PE. Are ACE inhibitors superior to beta blockers inpreventing endstage renal disease. retarding progressive renal function decline? JASN 1996; 7: 1400

(A) and Kidney Int; [Suppl ]: (in press)Gansevoort RT, de Zeeuw D, de Jong PE. Long-term benefits of

the antiproteinuric effect of ACE-inhibition in non-diabetic renalMethodsdisease. Am J Kidney Dis 1993; 2: 202–206In this prospective double-blind trial, 352 patients were classi- Apperloo AJ, de Zeeuw D, de Jong PE. Short-term antiproteinuric

fied according to baseline proteinuria (stratum 1, 1–3 g/24 h; response to antihypertensive therapy predicts long term GFRstratum 2, 3 g/24 h), and randomly assigned ramipril or pla- decline in patients with non-diabetic renal disease. Kidney Intcebo plus conventional antihypertensive therapy targeted at 1994; 45: [Suppl. 45]: S174–S178

The GISEN Group (Gruppo Italiano di Studi Epidemiologici inachieving diastolic blood pressure under 90 mmHg. TheNefrologia. Lancet 1997; 349: 1857–1863primary end-point was the rate of GFR decline. Analysis was

by intention to treat.

FindingsRational treatment in non-diabetic renal disease: what is the

At the second planned interim analysis, the difference in evidence?decline in GFR between the ramipril and placebo groups in

Eberhard Ritz, Ruperto Carola University, Heidelbergstratum 2 was highly significant (P=0.001). The independentadjudicating panel therefore decided to open the randomiz- There is controlled information available with respect to halt-

ing progression of renal failure on dietary protein restrictionation code and do the final analysis in this stratum (stratum 1continued in the trial ). Data (at least three GFR measure- and antihypertensive treatment. Although in animal experi-

ments [1] dietary protein restriction interfered with progres-ments including baseline) were available for 56 ramipril-assigned patients and 61 placebo-assigned patients. The sion, clinical use of dietary protein restriction has been

rediscovered only recently [2].decline in GFR per month was significantly lower in theramipril group than the placebo group (0.53 (0.08) vs Although in several short-term [3] or well-controlled stud-

ies of small sample size [4] a benefit was seen, a large0.88 (0.13) ml/min, P=0.03). Among the ramipril-assigned

Abstracts1078

controlled study, i.e. the MIDRD (Modification of Diet in organizing committee, Ab Donker of the Netherlands, HansBrunner of Switzerland, and David van Dijk of Israel as wellRenal Disease) study, failed to provide definite statistical

evidence for the efficacy of this therapy [5]. The MIDRD as the members of our advisory committee, Hans-HenrikParving, Eberhard Ritz, and Mike Steffes for their help andstudy can be criticized on several accounts, i.e. to short dura-

tion of observation and failure to consider acute haemo- advice.Before concluding I would like to express a certain concerndynamically mediated changes of GFR upon institution of a

low-protein diet. Any possible beneficial effect of protein for the patient, who has progressive renal failure. Over thepast day and a half we have heard the latest information onrestriction, if present [6 ], must be relatively small compared

to that of antihypertensive treatment. In children, a low- the pathogenesis and treatment of progressive renal failure.Several articles, describing the clinical approach to theseprotein diet was without any benefit [7].

The evidence is much more convincing for blood-pressure patients have been published. In fact their are clear guidelinesfor the treatment of the patient with diabetic nephropathylowering with conventional antihypertensive agents and more

recently for ACE inhibitors [8]. Administration of ACE [1–4] and the patient with hypertension [5,6 ]. Unfortunatelymost patients do not receive optimal treatment. Pommerinhibitors retards progression more effectively than can be

explained by blood-pressure lowering alone, at least in pro- et al. [7] found that of 66 diabetic patients, referred fornephrological advice, the creatinine clearance was belowteinuric patients. The beneficial effect on progression is

predicted by the initial decrease in proteinuria [9]. The select- 30 ml/min in 77% at time of referral. In spite of renal involve-ment, hypertension was poorly controlled in 97% and 75.%ive benefit of ACE inhibitors over alternative antihypertens-

ive agents is particularly pronounced when blood pressure is were not receiving angiotensin-converting enzyme inhibitors(ACE-I). The time interval from referral to dialysis was lesslowered little or not at all, while the antiproteinuric superior-

ity is progressively lost with increasingly efficacious lowering than 6 weeks in 52% of the patients. Thus this group ofpatients with advanced diabetic nephropathy had not beenof blood pressure [10]. A similar phenomenon was also

observed with respect to reaching renal end-points (doubling receiving optimal treatment and were referred to nephrolog-ists late in the course or their disease.of serum creatinine or renal replacement therapy) in the trial

on diabetic nephropathy (E. Lewis, personal communica- McClellan et al. [8] performed a retrospective analysis ofthe charts of a random selection of 587 patients admitted totion). The average rate of progression is significantly lower in

premenopausal women, while it is similar in men and in post- six hospitals, (two urban teaching hospitals), with a primaryor secondary diagnosis of diabetes mellitus or hypertension.menopausal women; this raises the issue of whether hormonal

replacement therapy should be considered in postmenopausal In less than 4% was any mention made in the history ofpossible renal disease. Less than 30% of these patients werewomen with renal disease.

Smoking has a proven deleterious effect on the appearance receiving ACE-I and over 10% were using NSAIDs. Screeningfor microalbuminuria in patients without a previous historyof proteinuria and progression of nephropathy in patients

with diabetes mellitus. The same holds true for patients with of renal disease was rare. and a urinalysis was performed inless than 60% of patients. About half of the diabetics and anon-diabetic renal disease [11].quarter of the hypertensives were found to have proteinuria

1. Brenner et al. N Engl J Med 1982; 307: 652 and almost one quarter of all patients had an elevated serum2. Maschio et al. Kidney Int 1982; 22: 371 creatinine. On discharge less than a third of patients were3. Rosman et al. Lancet 1984; 2: 1291 given ACE-I. The presence of proteinuria or an elevated4. D’Amico. Nephrol Dial Transplant 1994; 1590 serum creatinine had no influence on the use of ACE-I. In5. Klahr. N Engl J Med 1994; 330: 877

diabetic patients with proteinuria only 2% were given a low6. Maschio. Nephrol Dial Transplant 1997; 10: 1797protein diet, whereas NSAIDs were prescribed in 6%.7. Wingen. Lancet 1997; 349: 1117Mention of renal involvement was rare in the discharge sum-8. Maschio. N Engl J Med 1996; 334: 939

9. Aperloo. Kidney Int 1994; 45: S174 mary. The physicians treating this group of hospitalized10. Weidmann. Nephrol Dial Transplant 1993; 8: 582 patients did not show awareness of the possibility of renal11. Orth. J Am Soc Nephrol 1997; 8: 146 disease, did not screen for renal involvement and did not

prescribe the appropriate therapy. This meeting should be thestarting point for this effort.

Thus the guidelines for the care of the hypertensive anddiabetic patient, are not followed by a substantial proportion

Concluding remarks of primary care and hospital physicians. Clear guidelinesGeoffrey Boner; Institute of Hypertension and Kidney should be agreed upon for the care of the patient with non-Diseases, Rabin Medical Center, Beilinson Campus, Petah diabetic renal disease. Our efforts should then be aimed atTikva and Sackler Faculty of Medicine, Tel Aviv University, improving the care of patients with diabetic and non-diabeticTel Aviv, Israel renal disease by establishing special clinics and by the educa-

tion of the primary physician.When we started to organized this meeting over a year ago, Ihad hoped that I would be giving these remarks at the lowestpoint on the earth, i.e. at the Dead Sea. Following the ERA-

1. Viberti GC, Mogensen CE, Passa P, Bilous R, Mangili R. St.EDTA decision to move their congress to Geneva, we wereVincent Declaration, 1994. Guidelines for the prevention ofobliged to move this meeting to Switzerland. I am sure thatdiabetic renal failure. In: Mogensen CE (ed.) The Kidneyyou will all agree that the choice of the venue has been mostand Hypertension in Diabetes Mellitus, 2nd edn. Kluwer 1994;conducive to the study of progressive renal disease. I would551–527like to thank Michel Burnier, who agreed, at a very late date,

2. Clark CM, Lee DA. Prevention and treatment of the complica-to be responsible for the local arrangements. I must also stress tions of diabetes mellitus. N Engl J Med 1995; 332: 1210–1217the role of Carl Erik Mogensen, who was the moving force 3. Bennett PH, Haffner S, Kasiske BL, Keane WF, Mogensen CE,behind the organization of this meeting and without whom Parving H-H, Steffes MW, Striker GE. Screening and manage-we would not have been able to organize this outstanding ment of microalbuminuria in patients with diabetes mellitus:

recommendations to the scientific advisory board of the Nationalprogramme. I also wish to thank the other members of the

Abstracts 1079

Kidney Foundation from an ad hoc committee of the council on Group. National High Blood Pressure Education Programworking group report on hypertension in diabetes. Hypertensiondiabetes mellitus of the National Kidney Foundation. Am

J Kidney Dis 1995; 25: 107–112 1994; 23: 145–1587. Pommer W, Bressel F, Chen F, Molzahn M. There is room for4. Joint National Committee. The fifth report of the Joint National

Committee on detection, evaluation and treatment of high blood improvement of preterminal care in diabetic patients with end-stage renal failure. The epidemiological evidence in Germany.pressure (JNC V ). Arch Intern Med 1993; 153: 154–183

5. National High Blood Pressure Education Program. National Nephrol Dial Transplant 1997; 12: 1318–13208. McClellan WM, Knight DF, Karp H, Brown WW. Early detec-High Blood Pressure Education Program working group report

on hypertension and chronic renal failure. Arch Intern Med 1991; tion and treatment of renal disease in hospitalized diabetic andhypertensive patients: Important differences between practice and151: 1280–1287

6. National High Blood Pressure Education Program Working published guidelines. Am J Kidney Dis 1995; 25: 107–112