• No results found

Glomerular permselectivity in proteinuric patients after kidney transplantation


Academic year: 2020

Share "Glomerular permselectivity in proteinuric patients after kidney transplantation"


Loading.... (view fulltext now)

Full text


Glomerular permselectivity in proteinuric

patients after kidney transplantation.

R Oberbauer, … , A Schmidt, G Mayer

J Clin Invest.






To characterize the defect in glomerular permselectivity responsible for proteinuria after

renal transplantation, we studied 10 patients with moderate proteinuria (median 0.37 g/d,

range 0.20-0.79), 16 patients with the nephrotic syndrome (6.73 g/d, 3.9-14.6), 8 living

related donor transplant recipients without any history of rejection (median proteinuria 0.26

g/d, 0.06-0.58), and 12 healthy volunteers. The fractional clearance of neutral dextrans > 54

A was significantly higher in nephrotic patients, demonstrating a defect in glomerular size

selectivity. Using a log-normal model of glomerular pore size distribution, r*(5%) and r*(1%),

indices for the presence of large pores, were increased in the nephrotic patients. The

fractional clearance of negatively charged dextran sulfate was significantly higher in all

patient groups, indicating a loss of glomerular charge selectivity. Biopsy findings showed

more prominent glomerular lesions in the nephrotic group compared with the moderately

proteinuric group. We conclude that mild proteinuria late after renal transplantation is

associated with a defect in glomerular charge selectivity. The development of nephrotic

range proteinuria is associated also with a defect of glomerular size selectivity, which

correlates with prominent glomerular pathology.

Research Article



Permselectivity in Proteinuric Patients after Kidney Transplantation

Rainer Oberbauer, Martin Haas, Heinz


Ursula Bamas, Alice Schmidt, and Gert Mayer

Department of Internal Medicine III, Division of Nephrology and Dialysis; and *Department of Pathology, Division of Ultrastructural PathologyandCell Biology, University of Vienna, 1090 Vienna, Austria


To characterize the defect in glomerular


re-sponsible for proteinuria after renal transplantation, we studied10 patients withmoderate proteinuria (median 0.37

g/d, range 0.20-0.79), 16patients with thenephrotic

syn-drome (6.73 g/d, 3.9-14.6), 8 living related donor trans-plant recipients without any history of rejection (median

proteinuria 0.26 g/d, 0.06-0.58), and 12 healthyvolunteers. The fractional clearance of neutral dextrans >54 A was

significantly higher innephrotic patients, demonstrating a

defect in glomerular size selectivity. Using a log-normal model of glomerular pore size distribution, r*(5%) and r*(l%), indices for the presence of large pores, were in-creased in the nephrotic patients. Thefractional clearance

of negatively charged dextran sulfate was significantly higherinallpatientgroups,indicatingalossofglomerular charge selectivity. Biopsy findings showed moreprominent glomerular lesions in the nephrotic group compared with

the moderately


group. Weconclude that mild proteinuria late after renal transplantation is associated

withadefectinglomerular charge selectivity.The



isassociatedalso with

adefectofglomerular size selectivity, whichcorrelateswith prominent glomerular pathology. (J. Clin. Invest. 1995. 96:22-29.) Key words: kidney glomerulus * proteinuria . nephrosis*permeability . graft rejection


Withanincreasingnumber of renal

transplantations performed,

chronicallograftfailure isbecominga common causeof termi-nal retermi-nal failure. It remainstobedeterminedwhether

immuno-logicaland/ornonimmunologicalfactorsplaythemajorrole in theprogressionof this diseaseentity. Hypertensionand

protein-uriaareamong the clinical hallmarks of chronicallograft

dys-function (1, 2). However, theexact alterations of


function and structureresponsiblefor protein leakagein these


Ingeneral, the transport of macromolecules (andthus

pro-Addresscorrespondenceto GertMayer, M.D., Departmentof Internal MedicineHI,DivisionofNephrology, UniversityofVienna,


Gilrtel 18-20, 1090 Vienna, Austria. Phone: 140400-4390; FAX:


Receivedforpublication25May1994 andacceptedinrevisedform

14March 1995.

teins)through the glomerular filter is determined by their molec-ular size, charge, and shape (3). As endogenous proteins un-dergo a variable rate of tubular reabsorption they cannot be used directly to elucidate the characteristics of the glomerular filter.Therefore,tubularinert exogenous polymers are used for this purpose. Thesieving coefficient theta (9), the final urine

toplasma concentration ratioof these markers, divided by the urine to plasma concentration of an ideal marker for glomerular filtration isequivalenttothe Bowman's space to plasma concen-tration ratio. It isthereforethe most direct measure of the intrin-sic barrier properties ofthe glomerular capillary wall. Neutral dextrans of broad size distribution serve as markers for glomeru-larsize selectivity(4). Negatively charged dextran sulfate on the contrary is being usedtodetermineglomerular charge selec-tivity (5).

Weapplied these techniques to investigate glomerular perm-selectiveproperties inpatientsafter kidney transplantation with moderateand nephrotic range proteinuria in the late

posttrans-plant period. Furthermore, patients after living related donor

transplantation with moderateproteinuria, which never exhib-ited clinical evidence of (acute or chronic) rejection, and a group ofnormal,healthy controls were studied. In the nephrotic

patientsthe fractional clearances of albumin and IgG (as further estimates ofglomerularcharge and size selectivity)were deter-mined.

12 of the 16nephroticpatientsand 4 of the 10 moderately

proteinuric patientsunderwentallograft biopsy within 4 wk of the clearance studies. Biopsy cores were evaluated using the Banffclassification of renal allograft rejection (6) to gain

in-sightinto thestructural-functional relationshipofallograft pro-teinuria.



Group 1 consisted of 10 patients (3 male/7 female,mean age 52±5 yr)witha24-hurinaryprotein excretionbelow 1g/d (median protein-uria 0.37 g/d, range 0.20-0.79); group 2 consisted of 16 nephrotic

patients (10 male/6 female, mean age51±3 yr) withproteinuria in excessof 3.5g/d(6.73 g/d, 3.9-14.6) after cadavericrenal transplanta-tion;and group 3 consisted of 8 patients (5male/3female,meanage

26±2 yr)with proteinuria below 0.6 g/d (0.26 g/d, 0.06-0.58) who receivedalivingrelated donortransplant.As nativekidney biopsywas

performed only in aminority ofpatients, thediagnosis of the native

kidneydiseaseleadingto chronic renalinsufficiencywasmademainly

onclinicalimpressionswith chronicglomerulonephritisbeingthemost

frequentcause in allgroups(- 50%), followedbypolycystic kidney

disease and chronic interstitial nephritis. One patient became uremic because of diabeticnephropathy.Allpatientshadundergone

binephrec-tomy or had only minimal residual function oftheir native kidneys (i.e., urinary output <200 ml/d) before transplantation. 12 healthy

volunteers(7male/5female,meanage27±2 yr,group4) withoutany

history ofrenal disease or hypertension and with a negative urinary dipsticktestforalbumin servedascontrols.

22 Oberbauer, Haas, Regele, Barnas,Schmidt, andMayer J.Clin.Invest.

X The AmericanSocietyforClinicalInvestigation,Inc.

0021-9738/95/07/0022/08 $2.00


Table LGraft Function, 24-h UrinaryProtein Excretion(UPROTV)

6 mo beforeandatthe TimeoftheStudy andTimeafter Transplantationin the Three Patient Groups


Moderately related donor

proteinuric Nephrotic transplant patients patients recipients

(group1) (group2) (group3)

Serumcreatinine (Qmol/liter)

-6 mo 133±9 178±12 157±17

Atstudy 151+11* 208+15*t 157±17

UPRorV (g/d)

-6mo 0.31±0.06 5.10±1.10t 0.22±0.08

Atstudy 0.30±0.05 8.77tl.23*t 0.30±0.08

Time aftertransplantation

(mo) 40±9 58±9 56±10

P< 0.05 group 2vs.groups 1 and 3. *P< 0.05atstudyvs. -6 mo.Inthe6mopreceedingthestudy,allograft function remained stable onlyinlivingrelated donortransplantrecipients. Urinary protein

excre-tionwasprogressive andhigherinnephrotic patients when compared

with all other groups.

All allograft recipientswerein the lateposttransplant period, and patientsin group 1 and 2 hadaslowdeterioration ofexcretoryallograft

functionduringthe 6 moprecedingthestudy,but allhadserum creati-ninevalues below 310


(3.5 mg/dl).Noacuterejection had

tobe treatedforatleast6mobeforethepermselective studies and there

was nodifference between the two cadaver transplant groups in the total number ofacute rejection episodes during their posttransplant

course. Thepatients with the living donor transplantsneverexhibited

clinical evidenceofrejection. All patientswereinthe lateposttransplaht period, although group 1 patients were studied slightly earlier after transplantation. Details oftheclinicalcourseof the patientsarepresented


12/16patientsin group 2 and4/10patients ingroup 1 underwent

anallograft biopsy within4 wkof thestudy, andthebiopsycore was

scoredaccordingtotheBanff working classification forrenalallograft rejection (6).Insix nephrotic patients, materialwasavailable for


Basicimmunosuppressive therapy wassimilarin thethree groups andcanbeseeninTable II.CyclosporinAthroughlevelsweremeasured immediatelybefore theclearancestudies(i.e., 12 hafterthe lastdose).

To control blood pressure, 8 outofthe 10 patientsin group 1, 14of

the16patientsin group2,and 5ofthe 8living related donor transplant recipientswereonantihypertensive therapy.

One transplant patient refusedtoparticipatein the dextran sulfate

studiesand in anotheratechnical failureoccurredduringthegel perme-ationchromatography.Bothpatientswerein thenephroticgroup. Uri-naryIgG could only bemeasured in 14 outof 16 patients.

Functional studies

On the day ofthestudies thepatientsand controlsweretoldnot totake theirantihypertensivemedication.Before the clearancestudies, subjects

weretoldtovoidcompletelyandanoralwaterload of10ml/kg body

wtwas givento ensure ahigh urine output. Afterwards the subjects receivedabolus of Dextran 1000(Promitb;Kabi Pharmacia, Uppsala,

Sweden) (20 ml,0.15g/ml),followedbya 30-min infusion ofa 1:5 Dextran40/Dextran 70 mixtureat atotal doseof 130 mg Dextran/kg

bodywt. Dextran 40 and Dextran 70(Ebewe, Unterach, Austria) are

clinical preparations ofpolydisperse dextrans with mean molecular

weightsof40,000and70,000,respectively.Patientsorcontrols witha

Table II. Basic Immunosuppression in the Three Transplant Patient Groups


Moderately related donor

proteinuric Nephrotic transplant patients patients recipients (group 1) (group 2) (group 3)

Prednisolon (mg/d) 11±3 11±1 8±1

Azathioprin (mg/d) 67±8 64±7 82±7 CyclosporinAthrough level

(ng/ml) 106±5 93±5 94±11

Immunosuppressivetherapywasidentical in allpatientgroups.

historyofpenicillinallergywereexcluded from thestudytoeliminate

risk factors fordextranallergy andnoadverseeventsoccurredduring thestudy. We used this mixtureandnotthecommonlyused Dextran

40preparation alone inanefforttoextend thesievingcurve tohigher radii. Nonetheless,wewere notableto measuretheurinary

concentra-tionofdextrans>60 A.Aniothalamate(Conray 600; Cl Pharma, Linz, Austria) and para-aminohippuric acid (PAH)' bolus (Nephrotestt; BAG, Lich,Germany) followed byacontinuousinfusionwas adminis-teredimmediately after the Dextran 1000 infusion. Bolusamountand infusion rate wereadjustedto the estimated distribution volume and

glomerular filtrationrateinordertoobtain constant iothalamateplasma levels of- 1.5mg/dl andPAHplasmalevels of 2


Adextransulfate preparation (Ueno Fine Chemicals, Osaka,Japan) wastritiated(NewEnglandNuclearCorp.,Boston, MA), as a dose in

the toxicrangeof the unlabeled preparation would havetobeinfused

toachieveplasma concentrations within the assayablerange. Detailson

thepreparation usedaregiven elsewhere (7). The final preparationwas

checked for freeradioactivity by gel permeation chromatography(see

below). 300 sCiwasinfusedas abolus within 5minafter the neutral


Afteranequilibration period of 90min, twoclearance periods (30 mineach)wereperformed. Bloodwasdrawnatthebeginning and end of each period and analysis wasdone on pooled plasma. Urinewas

collectedthroughout the clearanceperiods and urinary fluid losseswere

replaced immediatelybyoral tapwater.

Analytical methods

Analysis of neutral dextran. Neutral dextran size separation of plasma

and urinesampleswasperformed usingasize exclusion HPLC(Jasco InternationalCo.Ltd., Tokyo,Japan)whichconsisted of aguard column

(Ultrahydrogel GuardColumn) andtwoserially arranged columns

(U1-trahydrogel 250 andUltrahydrogel 500;WatersChromatography,

Mil-ford, MA).The concentration ofthedextranfractions wasmeasured

usinganon-line refractiveindexdetector. The mobilephasewas a0.1 Mphosphate buffer,pH6.5.The columncalibrationstandards, which were a setof dextrans ofnarrowsize distributionwith molecular weights

determined by light scattering, were obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). Plasma andurine sample preparation

wasperformedas described in detailelsewhere (8).

We measured the clearancefor neutraldextrans over a size range from 30to60A.Below 30 A the accuracywaslimitedbyaninterfering

1.Abbreviations usedinthispaper:FSGS,focal andsegmental

glomer-ulosclerosis;Ap,effective filtration pressure;PAH,para-aminohippuric


plasma proteinpeak. Above 60 A determination of dextrans in the urine was notpossible due to very low concentrations. The recovery of neutral dextrans from urine was98±1%at30 A and declined to 92± 1.5% at

60A,the recovery from plasma at 30 A was 93±0.3% and declined to

84±1%at 60A.The interassaycoefficient of variation for the neutral dextran analysis was 3% at 30Aand6% at 60 A.

Analysis of dextran sulfate. The concentration of tritiated dextran

sulfateinurineandplasma wasmeasured by liquid scintillation counting (Beckman Instruments, Inc., Fullerton, CA). As reported previously, dextransulfate is significantlybound to plasma proteins (7, 8). Separa-tion of boundand free dextran sulfate was performed using gel

perme-ation chromatography. Each individual plasma and urine sample was put on topof a 1.5 x 30 cm Econo-column filled with a P-60 gel

(Bio-Rad Labs,Richmond, CA). PBS was used as rinsing solution and 23

fractions, 3 mleach,werecollected. Onepeak eluted infractions 5-7, another infractions 13-16; the recovery of total radioactivity in these two peaksfor all samples analyzed was 95±2%. The latter peak was

identifiedas dextran sulfateusing gel permeation chromatography of a

cold preparation ofthe compound. Thefractions obtained there were analyzed on a size exclusion HPLC with an on-line refractive index detector. The materialdetectedeluted at the sametimeasunfractionated colddextransulfate analyzed by HPLC immediately.When excess cold

dextransulfatewasadded toplasmasamples containing tritiated material and gel permeationwas performed afterwards, onlya single peak of tritiated dextran sulfateeluted. It wasconcludedthat thetritiated dextran sulfate wasreplacedbyunlabeled dextran sulfate. Asnosize standard

preparations areavailable for dextran sulfate, weindirectly estimated

the size of thefractionsby analyzing themon asize exclusionHPLC systemcalibratedwith neutraldextrans.The estimatedsize ofthe second

(free) peakwas 18-20 A and therefore corresponded well with the

manufacturer's statements.

Onlytheunboundplasma dextran sulfateconcentrationwasusedto

calculate the dextransulfate clearance. Using this procedure,the fraction

of non-protein-bound dextran sulfate was significantly lowerin

ne-phrotic patients (64±1%in group 2, 76±1% in group 1, 71±1% in group3, and 74±1% ingroup4;P<0.05group 2vs.group 1,group 3andgroup4).Dextransulfateis usedto removeplasmalipoproteins

inLDLapheresis(9).Itmight thereforebespeculatedthat inpatients

withhyperlipoproteinemia duetonephrosis moreof thecompound is proteinbound. When urinesampleswereanalyzed by gel chromatogra-phy, even in nephrotic patients a very small fraction (<2% oftotal

radioactivityin eachcase)eluted with thefirst peak.Therefore,the total

radioactivity ofthe urine sampleswas usedtocalculatethefractional

dextran sulfate clearance values.

Analysis ofPAHandiothalamate.Analysis ofPAHand iothalamate

wasperformedusingareversedphaseHPLC system(JascoInternational

Co.Ltd.)consisting ofaguardcolumn(Ultrasphere Octyl)andamain

column(Ultrasphere ODS;BeckmanInstruments, Inc.)asdescribedby

Morellietal.(10). The recovery of PAH inplasmawas92±2% and

in urine90±3%, for iothalamatethecorrespondingvalueswere93±2%

and92±2%. Theinterassay coefficient ofvariation for PAH measured in plasma and urine samples was 7 and 9% and for iothalamate 5 and3%,respectively. PAHclearance valueswereusedas amarker of

glomerular plasma flow. As the extraction of PAH is incomplete in advanced renalfailure,anextractionrateof 0.7wasassumed inpatients

withaGFR <40 ml/min(11).

Otheranalyses. Blood pressure wasmeasured bythe Riva Rocci method after thepatients had been restingfor 5 min. CyclosporinA

throughlevels were determinedusing aspecific monoclonal antibody


Incstar, Stillwater,MN). Protein concentrationswere de-terminedbyacolorimetricassay(ProteinAssay®;Bio-RadLabs). Con-centrations of albumin andIgGin serumand urineweremeasuredby radial immunodiffusion(lowerdetection limit50jHg/mlforplasmaand

20 pg/mlforurine). If the concentration of eitherproteinfell below the detection limit, a sensitive enzyme-linked immunosorbent assay (ELISA) with a lower detection limit of 3ng/mlwas used.


The fractional clearance theta(0)of any marker wascalculated using the


Theta marker(am) clearancemarker



Thesize selective properties of the glomerular filter were analyzed using two mathematical models. One is based on the assumption that themajor portion of the capillary wall is perforated by cylindrical pores of identical radius. Furthermore, there exists an additional parallel "shunt pathway" which does not discriminate on the basis of dextran size andthrough which a small fraction of the filtrate volume passes. The second model assumes that the radii of the pores follow a log-normal distribution. Bothmodels were described extensively by Deen and co-workers (12). It turned out that the log-normal distribution of poreradii best simulated the data obtained in vivo. Output variables of this model are the meanradiusofthe pores and the standard deviation of the mean radius (y and s) as well asr*(l%) and r*(5%).These latter two parameters determine the radius of the pores above which 1 and 5%of the glomerular filtrate areformed. In summary, these parame-terscharacterize the presence of large pores within the filter. Both mod-els require theeffective filtration pressure (Ap) as an input variable. As this pressure cannot be measured in humans, several assumptions have beenmade onAp in our study (variation ofApbetween 35 and 50 mmHg intransplant patients and 40 and 45 mmHg in healthy con-trols). These values for filtration pressure seem to be an acceptable

assumption accordingto valuesobtained with servonulling techniques inexperimental rodents. It should furthermore be pointed out that the variations inApmainly alter thetheoretical sieving of smaller dextran molecules and therefore do not influence the shunt parameter or r* values.

Theultrafiltration coefficient Kf was then calculated using different

ultrafiltrationpressurevalues. Statistics

All results are expressed as mean±SEM, except protein excretion,

which,becauseof itsnon-Gaussiandistribution, isgiven asmedian and range. Twotailed Student's ttestforpaired data and non-parametric Mann-Whitney test were used to assess differences between the two timepoints(6mobefore andatthetime of thestudy)withinonegroup. Data onall four groups wereanalyzed using ANOVAand Fisher'sexact

test. A P value< 0.05wasconsideredstatistically significant.

Thestudy wasapproved by the local human subject committee and awritten informedconsent wasobtained from everypatientbefore the



Excretory kidney function, renal plasma flow, and systemic blood pressure. Values formeanarterial blood pressure, iothala-mate,andPAHclearancesaregiveninTableIII.GFR andPAH clearance values in the controls areexpressedpersingle kidney

topermitcomparison. Alltransplantpatientgroupshad

compa-rable, although when compared with healthy individuals,

re-duced GFR values. However, only in nephrotic patients did

thisdifference reach statistical significance.Meanarterialblood pressure waselevated in transplantpatients, although again only nephroticpatients had significant hypertension.

Neutral dextran studiesandtheoreticalanalysis ofthe pore size distribution. The fractional clearance values for neutral

dextransin the nonnephrotic cadaver and livingrelated donor

transplant recipients were similar to, but uniformly elevated

above, theones obtained inhealthy controls. On thecontrary,

the fractional clearance for neutral dextrans >54 A was


Table III. MeanArterial BloodPressure(MAP), GFR,and PAH Clearance Values in the Three Transplant Patient Groups and in


Living Moderately related donor proteinuric Nephrotic transplant Healthy

patients patients recipients controls

(group 1) (group2) (group3) (group4)

MAP(mmHg) 108±4 111±3 106±5 100±2* GFR(ml/min) 47±6 40±7 48±7 63±8*t

PAHclearance(m/umin) 259±25 255±75 293±47 338±74* *P <0.05 group 4 vs. group 2. t Valueper onekidney. Bloodpressure wascomparable in the threetransplant patientgroups,althoughonly

nephrotic patients had significant hypertension whencomparedwith normal controls. GFR clearance valuesweresignificantly lowerin

ne-phrotic transplant recipients when compared withhealthycontrols, but

nodifferencewasfound between thegroups.

nificantly higher in transplant recipients with nephrotic range proteinuria than in any other group, demonstrating a defect in

glomerularsize selectivity (Fig. 1).

Table IV gives values for membrane parameters obtained using the log-normal model of Deenetal. (12). Apin normal control subjectswas set at40 and 45 mmHg and in recipients

was allowedtorange from 35to 50 mmHg. The lower value in each group was chosenaccordingto the calculated efferent arteriolaroncotic pressure, which averaged39 mmHg in

con-trols and33 mmHg intransplant recipients. Calculated values

formembrane pore parametersindicatedthat pore size distribu-tion was significantlyaltered in group 2recipients with heavy proteinuria.Inparticular, values for r*(5%) andr*(1%)inthis group exceeded those of the other groups, irrespective of the

assumptionsmaderegarding glomerular transcapillarypressure. As described by Remuzzi et al. (13), these two parameters determine the radius of the pores above which 1 and 5% of the glomerular filtrate are formed. Therefore, the finding of in-creased values for r*(5%) and r*(1%) irrespectiveof assumed


10 50 60 30


variation inAp indicates that the prevalence of large membrane pores was increased in group 2. Furthermore, the pore size distribution is wider in the nephrotic patient group. Table IV alsogives values for the whole kidney ultrafiltration Kf (for the healthy control group Kf is given per single kidney). As re-vealedby Table IV, calculated values for Kf depend heavily on assumed values for Ap. Formostassumed values ofAp, whole kidney Kf was reduced in the recipient groups. This reflects the reduction of GFR values in these groupstoless than the control single kidney value. Of note, values for Kf were lowest in group 2patients who exhibited the lowestmeanvalues for GFR despitehavingareduction inplasmaproteinconcentration and calculatedplasma oncotic pressure.

Dextran sulfate clearance studies.Results of clearance stud-ies using anionic dextran sulfate as a marker for glomerular chargeselectivity are shown in Fig. 2. The fractional clearance of dextran sulfate was significantly lower in normal control subjects when compared with all transplanted patients. There was, however, nodifference between the three patient groups

(fractional clearance ofnegatively charged dextran sulfate in healthy volunteers 0.57±0.04, 0.79±0.06 in the patients with moderate proteinuria, 0.76±0.05 in nephrotic patients, and 0.77±0.06 in living related donor transplant recipients). The fractional clearance of neutral dextrans of the size of our dextran sulfatepreparation has been showntobe closeto 1 (7). There-fore, avalue of 0.57 demonstrates intact charge selectivity in healthy patients, whereas values of 0.8 demonstrate a loss of glomerularcharge selectivity.

Fractionalclearance of albumin and IgG in nephrotic

pa-tients.Tofurthercharacterize thedefect that leadstoincreasing

proteinuria inallograft recipients,wedetermined the fractional clearance values ofIgGandalbumin in group 2 patients. The Einstein-Stoke's radius of albumin is 36 A, the radius of IgG

0 0

is estimated to be 55 A. Although apore radius of 50 A, as

estimated by neutral dextran sieving, probably overestimates

trueporesize, the passage of albumin through the glomerular filter should be hindered mostly by its negative molecular charge, whereas IgG can serve as an additional marker for glo-merular size selectivity. As can be seen from Fig. 3, the

frac-Figure 1. Fractional clearance of neutral

dex-trans(OD)of broad sizedistributionas a func-tion of molecular radius



frac-tionalclearance of neutral dextrans of





related donortransplant recipients (group 3,


wasuniformlyelevated but similartothat observed in controls (group 4, squares),


se-lectivity.(B) Cadaver renaltransplant

recipi-1%., entswith nephrotic rangeproteinuria(group

b 2, diamonds)exhibitadefect in glomerular sizeselectivityasevidentby significantly el-evatedfractional clearance values for neutral 4'0 5'0 60 dextrans >54Awhen




Table IV. Calculated Glomerular Pore Parameters Obtained Using a Log-Normal Model and Various Filtration Pressures

IL s r*(1 %) r*(5%) Kf AP

A A A A mL/min/mmHg mmHg

Moderatelyproteinuric patients (group 1) 54.17 1.13 76.6 70.4 10.76 35

58.63 1.09 72.4 68.5 4.74 40

59.33 1.08 71.5 68.1 3.12 45 59.55 1.07 71.2 68.0 2.33 50

Nephrotic patients(group2) 51.65 1.17 80.9 72.9 3.18 35

53.05 1.15 80.0 72.6 2.27 40

53.66 1.15 79.6 72.5 1.77 45

53.96 1.14 79.3 72.4 1.45 50

Livingrelateddonor transplantrecipients(group3) 49.54 1.17 78.3 70.4 8.35 35 54.54 1.12 74.6 69.1 4.36 40

55.85 1.11 73.4 68.6 2.98 45

56.34 1.10 72.9 68.3 2.27 50

Healthycontrols (group 4) 50.23 1.15 75.5 68.6 8.92* 40

53.82 1.12 73.3 67.9 5.26* 45

A,meanporeradius.s,standard deviation ofmeanpore radius.r*(1%and5%), radiusofpores, above which 1 and5%of the glomerular filtrate

isformed.Ap,filtrationpressure.Kf,glomerular ultrafiltration coefficient. *Value per onekidney. Calculated values for glomerularporeparameters

indicate thatthe poresize distributionwasalteredin group 2. Inparticular,valuesfors,r*(5%), andr*(1%)in this groupexceededthose in the other groups,irrespective of assumed variationinAp, indicating the increase of largemembrane pores innephrotic patients. Forall assumedvalues ofAp,Kfwasreducedin therecipientgroups,reflectingthe reductionof GFRvalues in these groupstoless than the controlsingle kidneyvalue.

tional clearance of albumin always exceeded the fractional clearance ofIgG.However,asproteinuria increased,the regres-sion line approached the line ofidentity (i.e.,


increased more rapidly, as did 9Albumin), indicating a more pronounced

defect ofglomerular size selectivitywithincreasingproteinuria. Histopathology ofthe transplant biopsies. Most biopsies showedarathercomplexpattern of chronic lesions. In all pa-tientsamoderatetosevere, mostlystriped, and sometimes dif-fuseinterstitial fibrosis could be observed. This was, together

with a hyalinotic arteriolopathy, attributed to chronic

Cy-closporin A toxicity. The main difference between the two

groups was found in glomerular pathology, as all moderately proteinuric patients showed only minor, nonspecific glomerular lesions.Innephroticpatients,glomerular pathologyinlight

mi-1- Figure 2.Fractional clearanceofdextran

sul-T T T fate


of narrow size

0,75- distribution. Patients

0,75- m | afterkidney

transplanta-S * tionexhibitadefect in


glomerular charge

selec-0,5- tivityasevidentbyan




0,25- charged dextran sulfate





proteinuric patients,

o-ssssz_ group_ , , l1,hatched




GROUP 1 GROUP 2 GROUP 3 GROUP4 nephrotic patients, group 2,openbar; living

re-lated donortransplantrecipients,group3,striped bar; healthy controls,

group4, closed bar). *P< 0.05 group 4vs.all other groups.

croscopywasevidentasmarked transplant glomerulopathy, glo-merular scaring, and signs of focal and segmental glomerulo-sclerosis (FSGS; focal and segmental or global sclerosis and obsolescence of the glomerular tuft, swollen or vacuolated, sometimes protein loadedpodocytes anddepositionofhyaline

material in sclerotic areas). In eight of the nephrotic patients theglomeruli showed changes typical for transplant







r2 = 0.814

p = 0.0001




/# 0



1E-05 0,0001 0, 001 O.01


Figure 3. Fractionalclearanceofalbumin





nephroticrenaltransplantpatients. Asproteinuria increases,the regres-sion lineapproaches theline ofidentity, indicatinganincreasingly

im-paired glomerularsizeselectivity.


Table V. Renal Histopathology Findings in Transplant Biopsies

Moderately proteinuric patients (group 1) Nephrotic patients (group 2)


n = 4 n= 12


Glomerular lesions

TX-glomerulopathy 0 0 0 1.3 8 1.9

Tubulointerstitial lesions

Fibrosis(andimflammation) 1.5 4 1.8 2.1 12 2.1

Tubular atrophy 1.3 3 1.7 1.8 12 1.8

Vascular lesions

Arteriolarhyalinosis 1.3 2 1.5 1 8 1.5

Fibrousintimalthickening 1.8 4 1.8 1.4 11 1.5


Scarredglomeruli 0 0 0 1.1 9 1.4

FSGS 0 0 0 1.3 9 1.8

n=O n=6


Foot processfusion 2.2 6 2.2

Electron densedeposits 0.3 2 1

Evaluation ofmorphological criteria forchronictransplant lesionsinlight microscopywasdoneaccordingtotheBanff classification(6). The lesionsweregraded0-3:(0= absent, 1 = slight,2= moderate,and 3 = severe). For electronmicroscopy thesamegradingstepswereapplied.

The scorefor the estimatedextentoffootprocessfusionwas:0 =9%, 1 = 10-29%, 2 =30-69%, 3 = 70%.MEAN(ALL),meanvalueof

scoresincludingallpatients ofthe group;n(POS),numberofpatients exhibitingagiven lesion (thus scoringwas atleast1);MEAN(POS),mean

value ofscores including only positive patients. The major difference betweenthetwogroupswasfoundinglomerular pathology, whichwas

present inlight microscopyexclusivelyinnephrotic patients. These patientsalsoexhibitedprominent footprocessfusion inelectronmicroscopy.

pathy (basement membrane doubling andthickening). Trans-plantglomerulopathy was accompanied in allcaseswith intimal fibrosis of the arterioles. In nine patients a sometimes very prominent glomerular lesion was FSGS. The termFSGS was

used as a description of morphological pattern rather than as anetiopathogenetic entity. It occurred either in association with otherglomerular lesionsor wasthe mainpathologic change in glomeruli. It is unlikely that the FSGS present in our patients

was recurrenceof disease(althoughexacthistological diagnosis of native kidney disease was notavailable in mostpatients),

asslowly increasing proteinuria developedlateafter transplanta-tion. In recurrent primary FSGS, nephrotic range proteinuria usually develops early afterengraftment (14).

Electron microscopywas available in 6 out of the 12

ne-phrotic patients and showed a widespread patchy or diffuse fusion andflatteningof thepodocytic footprocesses. Podocyte detachment from theglomerularbasement membrane, afinding often described inproteinuric experimental renaldisease, was also observed inourpatient population.However, the frequency of this alteration was very low. The results of the light and

electron microscopicevaluationsaresummarized in TableV.


The main aim of this study was to characterize thedefect in glomerular filter function that leads to proteinuria in renal

allo-graftrecipients late after transplantation. Several determinants ofglomerularpermselectivity (i.e., the capacity of the normal

glomerulustohinder the passageofmacromolecules from the blood compartmenttoBowman's space) have beendescribed,

themostimportant being glomerularsize andchargeselectivity. Largeand/ornegatively chargedmoleculesarelessreadily

fil-teredthan small and/or cationic molecules (3-5). In thisstudy

weused ideal markers forglomerularcharge and size

selectiv-ity,asboth neutral dextrans as well as dextran sulfate(in

con-trast toproteins)have been showntobe neither reabsorbednor

secretedby the renal tubular system(4, 15). After correction forwaterreabsorption, the concentration of these macromole-cules in thefinal urine represents the concentration in Bowman's space.

Dextransulfate,the marker forglomerular charge selectivity

in ourstudy,has been used mostly in animal studies (5, 15).

Recentlyit hasbeen demonstrated that dextran sulfate isprotein

bound in a significantmanner and therefore the technique of

calculating fractional clearance values has been modified

ac-cordingly(7, 8). Usingthesemodifications, Guaschetal. (7)

reported recentlythat inpatientswithmembranous nephropathy

fractional dextran sulfate clearance valueswereincreased when

compared with normal controls. This finding demonstrates a

defect inglomerular charge selectivity inthesepatients. Even

thoughthetechniqueofseparating proteinbound from unbound dextran sulfate wasslightlydifferent in theirstudy,ourresults

areremarkablysimilar. Normal controls demonstrateasieving

coefficient of dextran sulfate of0.6,indicatingapreserved


pro-teinuria havesignificantly higher fractional clearance of dextran sulfate (0.8). The same value is obtained in patients with mod-erateproteinuria irrespective of the origin of the transplant (ca-daver or living related donation) and nephrotic range protein-uria, suggesting that the defect inglomerular charge selectivity is the common alteration of glomerular permselectivity that leads to proteinuria after transplantation. The fractional clear-ance of large dextrans was increased in nephrotic transplant

recipientsascomparedwithall other groups. Theoretical analy-sis of thedextran sieving data using the log-normal model re-vealed that this change reflected an increase in the prevalence of large pores in the glomerular membrane. As in previous studies ofheavy proteinuria, this change was accompanied by a reduction in GFRdespiteareduction inplasma oncotic pressure, indicating a reduction in Kf associated with severe injury to the glomerular capillary wall. Incontrast tothe findings in group 2, dextran sieving curves were only slightly elevated across the size range studied inrecipientgroups 1 and3. The slight eleva-tion of the sievingcurve inthese groups is consistent with the

assumptionofadecrease in Apmediatedby cyclosporine(4). Itshould beemphasized, however, that the conclusions of this study regarding charge and size selectivity in transplant

recipi-ents are notdependentonparticular assumptionsregarding glo-merular pressure. In summary, in renal transplant recipients mildproteinuria is associated with a defect in glomerular charge selectivity. The development of nephrotic range proteinuria is associated also withadefect of glomerular size selectivity.

Asdemonstratedby our histopathological findings, the main structural correlate tothe defect in glomerular size selectivity

is thedevelopment ofglomerular pathology,as theindices for vascular and interstitial pathology are similar in the nephrotic andmoderately proteinuric cadavertransplant recipients.

Inter-estingly, on an electron microscopy level, detachment of the glomerular epithelial cells from the basement membrane wasa

veryinfrequent finding.This alteration has been shownto

corre-late with proteinuria in experimental renal disease (16). Of interesthowever,Milleretal. (17) have shown thatglomerular hypertrophy,inducedbyfour-fifths renalablation,increases the

areaofepithelial cell detachment inadriamycin nephrosis. On the contrary, Lafayette and co-workers (18) reported that

Cyclosporin A treatment reduces glomerular volume in rats.

Therefore,besides thepossibilityofasamplingerror,preserved podocyte adhesion tothe basement membrane in ourpatients mightbe duetoless prominentglomerular hypertrophy dueto treatmentwith CyclosporinA.

Thenephrotic transplant patientshad beentransplantedfor

a slightly longer time when compared with the nonnephrotic


eval-uated proteinuria in the nephrotic patients 18 mo before the

studytomatch the timeafter transplant between thetwogroups.

At thattimepoint proteinuriawas1.25g/d,avaluequite similar

to that observed in our moderately proteinuric groups at the time of the study. Thisalso is in line withtheassumptionthat

proteinuria in these patients develops by an initial defect in

glomerular charge selectivity. Later on, as heavy proteinuria

develops, glomerularsizeselectivitybecomesincreasingly

im-paired. Even in nephrotic patients, this sequence ofevents is prolonged as shown by ourdata onthe relationship between thefractional albumin and IgG clearance.

Theresultsweobtained inourtransplantpatientsaresimilar

to that obtainedby others in native kidney disease.Inpatients

with diabetic nephropathy, Deckertetal. (19) recently reported

a similar sequence in the evolution of proteinuria. Using an IgG/IgG4selectivityindex as a marker for glomerular charge

selectivity,adefectwasevident inpatients with albuminuria in the range of 100-300 mg/24 h. Neutral dextran sieving was only altered inpatientswith albuminuria of> 1,500 mg/d ( 19). Scandling and co-workers (20) found a similar relationship between the fractional clearance of albumin and IgG in patients with variousnephrotic native kidney diseases.

With the increasing number of transplants performed, chronic allograft dysfunction is becoming a major problem in clinicalnephrology. If proteinuria in patients late after kidney

transplantation is causedbyacomparable defect asnephrosis in native kidney disease, similar treatment options might be available. In nativekidney disease, blockadeof the renin angio-tensin system has been shownto effectively lower proteinuria

(21, 22).Usingneutraldextransieving analysis,several authors

wereabletodemonstrateabeneficial effectof angiotensin

con-vertingenzyme blockadeonglomerularsize selectivity in these

patients (10, 13, 23, 24). Accordingly, Traindl et al. (25)

re-portedabeneficialeffect of angiotensinconverting enzyme in-hibitor treatment on proteinuria in renal transplant recipients.

Furthermore, Remuzzi et al. (26) suggested recently that

proteinuria persemight lead to a progressive decline in renal excretory function. It remains to be determined, however, whetherareductionin proteinuria in renal transplant recipients

mightslow theprogression ofchronic transplant failure.


Thisstudywassupported bytheOsterreichischerFond zurForderung

der Wissenschaftlichen Forschung, grant P9002MED.


1.Cheigh,J.S.,R. H.Haschemeyer,J.C. L.Wang,R. R.Riggio,L.Tapia, K. H.Stenzel,and A. L.Rubin.1989.Hypertensioninkidney transplant recipients: effectonlong-termrenalallograftsurvival. Am. J.Hyperten.2:341-348.

2. Harlan,W.R., K. R.Holden,G. M.Williams, andD. M.Hume. 1967. Proteinuria andnephrotic syndromeassociatedwithchronicrejection of kidney transplants.N.Engl.J.Med. 277:769-776.

3.Deen,W.M.,C. R.Bridges,andB. M. Brenner.1983.Biophysicalbasis of glomerularpermselectivity.J.Membr.Biol. 71:1-10.

4.Chang,R.L.,I. F.Ueki,J. L.Troy, W. M.Deen,C. H. R.Robertson, and B. M. Brenner. 1975.Permselectivity of theglomerular capillarywall to macromolecules.II. Experimentalstudies in ratsusingneutraldextran. Biophys. J. 15:887-905.

5. Chang, R. L., W. M. Deen,C. H. R. Robertson, and B. M. Brenner. 1975.Permselectivityof theglomerular capillarywall.m.Restricted transport of polyanions. KidneyInt. 8:212-218.

6. Solez, K.,R. A. Axelsen,H.Benediktson,J. F. Burdick,A. H.Cohen, R. B.Colvin,B. P.Croker,D.Droz, M. S.Dunnill,P. F.Halloran,etal.1993. Internalstandardisation of criteria for thehistologic diagnosisof renalallograft rejection:theBanffworking classificationofkidney transplant pathology. Kidney Int.44:411-422.

7.Guasch, A.,W. M.Deen,and B. D.Myers.1993.Chargeselectivityof the glomerularfilter barrier inhealthyandnephrotichumans. J.Clin. Invest. 92:2274-2282.

8.Mayer,G.,R. A.Lafayette,J.Oliver,W. M.Deen,B. D.Myers,and T. W. Meyer. 1993. Effects ofangiotensinII receptor blockadeon remnantglomerular permselectivity.KidneyInt. 43:346-353.

9.Yokoyama, S.,R.Hayashi,M.Satani,and A. Yamamoto. 1985. Selective removalof lowdensity lipoprotein by plasmapheresisin familial hypercholesterol-emia.Arteriosclerosis.5:613-616.

10.Morelli, E.,N.Loon,T.Meyer,W.Peters,and B. D.Myers.1990. Effects ofconvertingenzyme inhibitiononbarrier function indiabeticglomerulopathy.


11.Golbetz,H., V. Black,0.Shemesh, and B. D.Myers.1989.Mechanism


of the antiproteinuric effect ofindomethacin innephrotichumans.Am. J.Physiol. 256:F44-F51.

12. Deen,W.M., C. H. R. Bridges, B. M. Brenner, and B. D. Myers. 1985. Heteroporous model ofglomerularsize selectivity: application to normal and nephrotic humans.Am.J.Physiol.249:F374-F389.

13. Remuzzi, A., E.Perticucci,P.Ruggentini,L.Mosconi,M.Limonta, and G.Remuzzi.1991.Angiotensin convertingenzymeinhibitionimproves glomeru-lar size selectivity in IgAnephropathy. KidneyInt.39:1267-1273.

14. Artero, M., C. Biava,W.Amend, S.Tomlanovich, and F. Vincentini. 1992. Recurrent focalglomerulosclerosis:naturalhistoryandresponsetotherapy. Am.J.Med.92:375-383.

15. Bennet, C.M., R. J.Glassock,R. L. S.Chang,W. M.Deen, C. H. R. Robertson,and B. M. Brenner.1976. Permselectivity oftheglomerular capillary wall:studiesofexperimental glomerulonephritisin the ratusingdextransulfate. J.Clin.Invest. 57:1287-1294.

16.Whiteside, C.,K.Prutis,R.Cameron,and J.Thompson. 1989. Glomerular epithelial detachment,notreducedcharge density, correlateswithproteinuriain adriamycin and puromycin nephrosis. Lab.Invest.61:650-659.

17.Miller,P. L.,J.Scholey,H.Rennke,and T. Meyer. 1990. Glomerular hypertrophyaggravates epithelialcell injury innephroticrats. J. Clin. Invest. 85:1119-1126.

18. Lafayette,R. A., G.Mayer, and T. Meyer. 1993.The effect ofblood pressure reduction oncyclosporine nephrotoxicityintherat. J. Am. Soc.NephroL 3:1892-1899.

19. Deckert, T.,A.Kofoed-Enevoldsen, P. Vidal,K.Norgaard, H.B. An-dreasen, and B. Feldt-Rasmussen. 1993. Size and charge selectivity of glomerular filtration in Type1(insulin dependent) diabetic patients with and without albu-minuria. Diabetologia. 36:244-251.

20. Scandling,J. D., V. M. Black, W. M. Deen, and B. D. Myers. 1992. Glomerular permselectivity in healthy and nephrotic humans. Adv. Nephrol. NeckerHosp. 21:159-176.

21.Taguma, Y., Y. Kitamoto,G.Futaki, H. Ueda, H.Monma,M.Ishazaki, H. Sekino, and Y. Sasaki. 1985.Effect of captopril on heavy proteinuriain azotemic diabetes. N.Engl.J.Med. 313:1617-1620.

22. Heeg, J. E., P. E. De Jong,G.K. Vander Hem, and D. De Zeeuw. 1987. Reduction of proteinuria byangiotensin convertingenzymeinhibition. Kidney Int.32:78-83.

23. Remuzzi, A., P.Ruggentini, L.Mosconi, V. Pata,G. Viberti, and G. Remuzzi. 1993. Effectof low doseenalapril onglomerularsize selectivityin humandiabetic nephropathy.J. Nephrol. 6:36-43.

24. Thomas, D. M., A. N.Hillis, G.A.Coles,M.Davies, and J.D.Williams. 1991. Enalaprilcan treatthe proteinuria ofmembranousglomerulonephritis with-outdetriment tosystemicorrenalhemodynamics.Am.J.Kidney.Dis.18:38-43. 25.Traindl, O.,S.Falger, S. Reading,M.Banyai,B.Liebisch, J. Gisinger, E. Templ, G. Mayer, andJ. Kovarik. 1993. The effectsoflisinoprilonrenal function inproteinuricrenal transplant recipients. Transplantation(Baltimore). 55:1309-1313.


Related documents

This Service Level Agreement (SLA or Agreement) document describes the general scope and nature of the services the Company will provide in relation to the System Software (RMS

q w e r t y Description Rod cover Head cover Cylinder tube Piston rod Piston Bushing Cushion valve Snap ring Tie rod Tie rod nut Wear rod Rod end nut Back up O ring Rod seal Piston

This is the recurring motto of the unedited treatise Diez privilegios para mujeres preñadas 4 (Ten Privileges for Pregnant Women), written in 1606 by the Spanish physician

In answer to these questions, this paper offers two main reasons for the relative neglect of this motivational component: the first is related to the histori- cal roots of the

Named in memory of Conall 6 Fearraigh, a Donegal singer, it was for amhranafocht gan tionlacan nach sean-nos f (unaccompanied singing which is not sean-nos). This was meant to

Consultation with a wide variety of potential stakeholders, including the public, NHS health and information technology professionals, government departments, the Human

ANGIO-IMMUNOBLASTIC lymphadenopathy (AILD) - a case report by 0 Azizon, NH Hamidah, 0 Ainoon, SK Cheong and KS Phang (abstract).. ANTIBODY responses of dengue fever

Pulverization of the limestone followed by wet separation using dispersion cum settling technique leads to liberation and separation of the clay minerals present in it leading to