216

Stretching of Glomerular Afferent Arteriolesin the Swollen Renal Allograft Undergoing

Acute Rejection

CARIN L. ALLHISER, JOHN E. STEFANIAK, LEE A. HEBERT, ANTHONY J. LAPORTA,

ARTHUR L. RILEY, AND DEREK SAMPSON

SUMMARY Marked renal swelling occurs in acute renal allograft rejection. The impaired renalfunction in this state could be related in part to mechanical adjustments that the renal tubular andvascular network must make as they participate in the overall expansion of renal parenchyma! volume.In the present study we examined whether vascular stretching occurs in the swollen acutely rejectingkidney. From each of five pairs of dogs matched for weight, a donor animal was chosen from which acontrol kidney was removed, weighed, and perfused at constant pressure (150 mm Hg) with polymer-izing silicone rubber. The contralateral kidney was transplanted into the recipient dog which was thenbilaterally nephrectomized. On the sixth posttransplant day, the transplant kidney was removed,weighed, and perfused with silicone rubber, as above. Multiple coronal sections were made of eachcontrol and transplant kidney and a total of 240 photomicrographs were taken of the coronal sectionsfrom unselected fields of outer and inner cortex. The photographic slides were then coded, randomized,and projected. Measurements were then made, from the projected images, of the glomerular afferentarteriolar length (AL), and width (AW), and the glomerular width (GW). The code was then broken andthe measurements in the control and transplant kidneys collated. We found that transplant AL wassignificantly longer (mean ratio: transplant/control •= 1.32 ± 0.06, P < 0.001) and transplant AWsignificantly narrower (mean ratio: transplant/control - 0.88 ± 0.04, P< 0.05). We also found that GWwas slightly but significantly decreased in the transplant compared to the control kidney. We concludethat afferent arteriolar stretching occurs in swollen renal allografts and accounts, at least in part, forthe impaired renal function observed in acutely rejecting renal allografts. Ore Res 44: 216-222, 1979

MARKED expansion of renal parenchymal volumeabove normal commonly occurs in states of acuterenal injury, such as ischemic damage or transplantrejection (Abrams, 1972; Fletcher et al., 1969). Inprevious studies, we provided indirect evidence thatthe renal swelling in these states is due, principally,to an increase in compliance of the kidney (Hebertet al., 1975) and that compliance-mediated (as op-posed to pressure-mediated) expansion of renal pa-renchymal volume may impair renal function (He-bert et al., 1975, 1978 in press). One mechanism bywhich compliance-mediated expansion of renal pa-renchymal volume could affect renal function isthat the renal vasculature might become stretched,as the vascular network participates in the overallexpansion of renal parenchymal volume. In thepresent study we assessed this possibility by com-paring glomerular afferent arteriole dimensions incontrol and transplant kidneys. This vascular seg-ment was chosen because its anatomic limits can bereadily defined and measured and because it is a

From the Departments of Medicine and Surgery and the Allen BradleyMedical Sciences Laboratory of the Medical College of Wisconsin, andthe Milwaukee County General and the Wood Veterans AdministrationHospitals, 8700 West Wisconsin Avenue, Milwaukee, Wisconsin.

Supported by National Institutes of Health Grant HL13323.Reprint requests to Lee A. Hebert, M.D., Renal Section, Department

of Medicine, Medical College of Wisconsin and Milwaukee County Med-ical Complex, 8700 West Wisconsin Avenue, Milwaukee, Wisconsin 53226.

Received May 1, 1978; accepted for publication August 30, 1978.

major resistor in the renal vascular circuit (Abe etal., 1970).

Methods

Surgical PreparationFive pairs of male mongrel dogs, each pair

matched for weight, were studied. Each pair wasanesthetized with intravenous sodium pentobarbi-tal (30 mg/kg), and endotracheal tubes were in-serted and connected to a constant volume, posi-tive-pressure ventilator which was adjusted accord-ing to a nomogram to achieve normal levels ofventilation. From each pair of dogs, a donor and arecipient were selected arbitrarily and were pre-pared for surgery under sterile conditions. In thedonor dog, both kidneys were freed of their non-hilar perirenal attachments and either the right orleft kidney was chosen as the control kidney. Thedog was rejected as a donor if the kidneys appeareddifferent in size. The hilar vessels of the controlkidney then were double-clamped and incised be-tween the clamps. The excised kidney was weighedand prepared for silicone rubber infusion (Microfil,Canton Bio-Medical Products). The contralateralkidney (transplant kidney) was excised in a similarfashion and perfused with iced, heparinizedRinger's-lactate solution until the venous effluentwas clear (about 250 ml). This kidney then was

by guest on May 16, 2018

http://circres.ahajournals.org/D

ownloaded from

GLOMERULAR AFFERENT ARTERIOLAR STRETCHING/Allhiser et al. 217

transplanted into the iliac fossa of the recipient dogwith vascular anastomoses to the iliac vessels andwith the ureter connected to the bladder. The re-cipient's native kidneys were then removed and theabdominal incision closed. During the surgical pro-cedure, the donor animal received intravenously0.5-1.0 liter of Ringer's lactate and the recipientanimal received intravenously 2 liters of Ringer'slactate. The serum creatinine was measured in thedonor and the recipient on the day of kidney trans-plantation and daily in the recipient animal, there-after. On the 6th posttransplant day, the recipientdog was anesthetized and the renal allograft re-moved (as described above), weighed, and preparedfor silicone rubber infusion.

Silicone Rubber InfusionSilicone rubber was prepared in a 4:5 ratio of

silicone rubber to diluent with 5% (by volume) ofcatalyst added to cause polymerization. The siliconerubber was infused via a canula placed in the renalartery within 3 minutes after nephrectomy. Theinfusion was controlled by a constant infusion pumpwhich was continually adjusted to maintain theinfusion pressure at about 150 mm Hg. The infusionpressure was monitored by means of a Stathampressure transducer which was attached via a sidearm to the renal artery catheter.

Silicone rubber infusion was continued until fill-ing of superficial cortical vessels and/or effux ofsilicone rubber from the renal vein was noted. Theinfusion then was stopped, the hilar structuresclamped, and the kidney stored at 4°C for 24 hoursto allow the silicone rubber to harden.

Preparation and Photography of SiliconeRubber-Injected Specimens of Kidney

The kidney injected with silicone rubber was cutinto coronal sections about 0.5 cm thick and placedin alcohol to remove water from the specimen. Thiswas done by placing the specimens in 25% ethylalcohol on the first day and then increasing theconcentration of alcohol to 50% on the second day,75% on the third day, 95% on the fourth day, andabsolute alcohol on the fifth day. On the sixth day,the specimens were placed in methyl salicylate forclearing. The specimens were photographed, whilein the methyl salicylate solution, through a Leitzmicroscope using a 3.5 magnification objective. APentex-Asahi 35 mm camera with Kodak Pana-tomic X film was used.

Filling of the transplant kidney vasculature withthe silicone rubber tended to be patchy comparedto fining observed in the control kidney. Thus, toinsure that the photographs of the control kidneycortex corresponded to areas photographed in thetransplant kidney cortex, the transplant kidney wasphotographed first. The corresponding areas in thecontrol kidney cortex then were photographed.Photographs were taken first of the inner cortexclosest to the hilum and then, moving circumfer-

entially about 4 cm at a time, until the entire innercortex was photographed. The outer cortex thenwas photographed in the same fashion. There wasno selection of the fields, except that prior to thetaking of each photograph, the field was inspectedto determine if the vasculature in that field wasfilled with silicone rubber. If any filling with siliconerubber was present, the field was photographed. Ifno vascular filling with silicone rubber was present,the field was moved circumferentially about 4 mmat a time until a field with vascular filling withsilicone rubber was found.

Analysis of PhotographsThe photographic negatives of the kidney speci-

mens were set in conventional 35-mm mountings,numbered, randomized, and then placed in a car-ousel by a person who was not involved in theinterpretation of the slides. The slides then wereprojected onto a screen 277 cm from the projectorlens. To determine the degree of magnification ofthe projected image, a hemocytometer grid withlines 0.2 mm apart was photographed with the samearrangement used to photograph the kidney speci-mens. The distance between the projected grid lineswas 40.0 mm. Thus, the total magnification of theprojected image was 200 X.

From each photographic slide of the tissue spec-imens, the following determinations were madefrom the projected images.

Glomerular Afferent Arteriolar Length (AL)The length of each glomerular afferent arteriole,

with a visible origin and connection with a glomer-ulus, was measured from the origin of the arterioleon its feeding artery to its glomerular termination(see Fig. 1). This was done with a hand-held devicewhich measured, in millimeters the distance a ro-tating wheel moves in tracing the measured dis-tance (Selsi map measurer, West Germany). Themeasurement of AL tends to underestimate the trueglomerular afferent arteriolar length for two rea-

Afferent Arteriole

Glomerulus

Artery

FIGURE 1 Schematic representation of the microana-tomic measurements made in this study. AL = afferentarteriolar length; GD = glomerular distance; GW =glomerular width. In addition, we measured afferentarteriolar width (A W) at the midpoint of all arterioleswhich were in focus.

by guest on May 16, 2018

http://circres.ahajournals.org/D

ownloaded from

218 CIRCULATION RESEARCH VOL. 44, No. 2, FEBRUARY 1979

sons: (1) If the glomerulus is bent back upon theafferent arteriole or if the afferent arteriole origi-nates from the anterior or posterior surface of thefeeding artery, some of the afferent arteriole will behidden from view. (2) Any deviation of the glomer-ular afferent arteriole from the plane of the photo-graph will result in an underestimate of the trueafferent arteriole length.

Glomerular Distance (GD)

For each glomerulus in which AL was measured,the shortest distance from the feeding artery to theglomerulus was also measured (see Fig. 1). GD wasmeasured to assess the degree of straightness of theafferent arteriole. For example, if we found that themean GD were only slightly less than the mean AL,this would indicate that most afferent arterioles arerelatively straight.

Glomerular Afferent Arteriolar Width (A W)The width of each afferent arteriole, which was

judged to be in focus at its midpoint, was measuredat its midpoint, using a ruler calibrated in 1/32 ofan inch (0.794 mm). Good focus is especially impor-tant for the satisfactory measurement of short dis-tances, such as AW.

Glomerular Width (GW)

For each glomerulus which was filled completelyfrom side to side (as assessed by a smooth, roundedcontour of the injected glomerulus), the widest,visible dimension of the glomerulus (see Fig. 1) wasmeasured. GW tends to underestimate the trueglomerular width since, for a given glomerulus, thephotograph may not show the glomerulus in itswidest dimension. GW was measured to assesswhether any stretching in renal vasculature mightbe accompanied by a change in glomerular size.

ResultsAll mean values are shown ± one standard error

of the mean.

Gross Anatomic FindingsThe mean control kidney weight was 55.2 ± 5.5

g. These kidneys were infused with an average of5.2 ± 0.4 ml of silicone rubber at the time whenfilling of the superficial cortical vessels became ev-ident and the infusion was stopped. In all controlkidneys there was good filling of both outer andinner cortex as typified in Figure 2.

By the sixth posttransplant day, the transplantkidneys had increased significantly in weight. Themean transplant kidney weight at the time of trans-plant nephrectomy was 131.5 ± 11.0 g (P < 0.01,compared to respective control kidneys by paired t-test). A significantly larger volume of silicone rub-ber (11.5 ± 0.7 ml, P < 0.01 compared to respectivecontrol kidneys) was required to inject the trans-

plant kidneys to achieve filling of superficial cortexvessel or to detect the appearance of silicone rubberin the venous effluent. This occurred despite thefact that much of the outer cortex of the transplantkidney did not fill with the injected silicone rubber(Fig. 2).

Microanatomic FindingsFigure 3 shows selected photomicrographic fields

of silicone rubber-injected control and transplantkidneys. Note that in the control kidneys the glo-meruli tend to lie close to their feeding artery andare connected to the feeding artery by a relativelyshort and sometimes gracefully arched afferent ar-teriole. By contrast, in the transplant kidneys theglomeruli appear to be pulled away from their feed-ing artery and some of the afferent arterioles appearto be stretched and narrowed.

Table 1 compares the microanatomic measure-ments of the control kidneys to that of the trans-plant kidneys. The actual dimensions of the renalstructures are shown in micrometers. As can beseen, transplant kidney AL is significantly longerthan control kidney AL, in both outer and innercortex. This was shown by comparing, by unpairedf-test, the mean value for AL from all control kidneymeasurements to the mean value for AL from alltransplant kidney measurements. In addition, foreach dog we calculated the ratio (mean AL, trans-plant kidney/mean AL, control kidney), for bothouter and inner cortex. Each ratio was greater than1.00 (range, 1.06-1.60). There was no differencebetween the mean ratios for outer and inner cortex;thus the ratios for outer and inner cortex werecombined, averaged, and this mean ratio comparedto 1.00, as shown in Table 1 (bottom line). Thisanalysis also shows that the AL of the transplantkidneys is significantly greater than AL of the con-trol kidneys. The data for GD were analyzed in thesame way as the data for AL. As can be seen fromTable 1, the interpretation of the GD data is thesame as the interpretation of the AL data.

The finding of increased afferent arteriolar lengthand greater distance between the glomerulus andits feeding artery is largely the result of stretching(an increase in true length) of the glomerular affer-ent arteriole. That is, it seems unlikely that theincrease in AL and GD are the result simply ofuncoiling of the afferent arterioles (an increase inafferent arteriolar length visible to the camera, butno increase in the true length of the afferent arter-iole). This interpretation follows from the fact that,as depicted in Figure 3, most afferent arterioles innormal kidneys are straight or nearly straight andnone is highly coiled. This interpretation receivesquantitative support by examining the relationshipbetween AL and GD, shown in Table 1. As can beseen, AL and GD are nearly equal, indicating thaton the average, glomerular afferent arterioles followa nearly straight path from the feeding artery to the

by guest on May 16, 2018

http://circres.ahajournals.org/D

ownloaded from

GLOMERULAR AFFERENT ARTERIOLAR STRETCHING/Allhiser et al. 219

FIGURE 2 Representative sections of silicone rubber-injected control and transplant kidneys. In the control kidney,note the abundant vascular filling of all portions of cortex. By contrast, in the transplant kidney, note the patchyfilling of cortex and the better filling of the polar cortex (right side of figure) and medial cortex (right side and bottomof figure) compared to the lateral cortex (top of figure).

glomerulus. Thus, uncoiling of the glomerular affer-ent arteriole cannot account for the large increasesin AL observed in the transplant kidney (Table 1).

It also seems unlikely that any systematic errorin the selection of glomerular afferent arterioles formeasurement can account for the finding of greaterAL and GD in transplant vs. control kidneys. Infact, if anything, the techniques used tend to un-derestimate the increase in AL and GD in trans-plant kidneys. This interpretation is based on thefact that control kidneys have more short glomer-ular afferent arterioles, and short afferent arteriolesare difficult to find, particularly in control kidneys,because of the tendency to more abundant vascularfilling with silicone rubber (Fig. 3). Thus, in thecontrol kidney data for AL and GD, there is under-representation of glomeruli with short afferent ar-terioles because these afferent arterioles are diffi-cult to identify, and thus often were not measured.Consequently, the mean control kidney AL shownin Table 1 is probably greater than the true meanAL for control kidneys. The greater difficulty in

finding measurable glomerular afferent arterioles inthe material obtained from control kidneys alsoaccounts for the fact that there are fewer measure-ments in the control kidneys, compared to thetransplant kidneys, as shown in Table 1.

The data on AL are also of interest because theyshow that in the control kidney the average AL ofinner cortex is significantly greater than the averageAL of outer cortex (P < 0.001).

The GW data shown in Table 1 confirm previousobservations that the glomeruli of inner cortex arelarger than the glomeruli of outer cortex (Schneideret al., 1972). We also found that the glomerulitended to be smaller in the transplant kidneys com-pared to the control kidneys. Although this couldnot be shown by unpaired /-testing of the controlkidney data vs. the transplant kidney data, it wasshown by determining for each kidney the ratio(GW, transplant kidney/GW, control kidney) forboth outer and inner cortex. When these ratios werepooled, the mean ratio was significantly less than1.00 (0.97 ± 0.009, P < 0.02). The finding of a

by guest on May 16, 2018

http://circres.ahajournals.org/D

ownloaded from

220 CIRCULATION RESEARCH VOL. 44, No. 2, FEBRUARY 1979

Control Kidney-Outer Cortex Transplant Kidney-Outer Cortex

Control Kidney-Inner Cortex Transplant Kidney-Inner Corte

Control Kidney-Inner Cortex Transplant Kidney-Outer Cortex

FIGURE 3 Photomicrographs of exemplary cortical fields in silicone rubber-injected control and transplant kidneys.In the control kidney specimens, the arrows identify individual or groups of afferent arterioles which demonstrate thatmany afferent arterioles are slightly curved, or pursue a mildly undulating course. In addition, in the left lower panel,note the difficulty in clearly identifying the origin and termination of short afferent arterioles in fields which haveabundant vascular filling with silicone rubber. In the transplant kidneys, the arrows identify individual or groups ofafferent arterioles which demonstrate that many afferent arterioles have a "straight as a string" appearance,suggesting stretching of the afferent arterioles. Also, most of the glomeruli appear to lie at a distance from theirrespective arteries, making the origin and termination of most afferent arterioles relatively easy to identify.

tendency for smaller glomeruli in the transplantkidneys is consistent with the hypothesis that theglomeruli was underperfused because of an increasein afferent arteriolar resistance.

Table 1 also shows that the afferent arterioles ofthe transplant kidney are narrowed compared to

those of the control kidney. This was shown bycomparing, by unpaired f-test, the mean value forAW from all control kidney measurements to themean value for AW from all transplant kidneymeasurements. In addition, for each dog we calcu-lated the ratio (mean AW, transplant kidney/mean

by guest on May 16, 2018

http://circres.ahajournals.org/D

ownloaded from

GLOMERULAR AFFERENT ARTERIOLAR STRETCHING/A Uhiser et al. 221

TABLE 1 Microanatomic Measurements in Control Kidneys (CK) and Transplant Kidneys (TK)

Outer cortexControl kidneysn = 91 (28)Transplant kidneysn = 211 (39)

Inner cortexControl kidneysn = 212 (76)Transplant kidneysn - 371 (97)

Mean of the mean ratio from eachexperiment. Data from outer andinner cortex pooled.

Glomerular affer-ent arteriole

length (AL, JIM)

172 ± 11

248 ± 8P< 0.001*

236 ± 9

288±9P< 0.001*

(AL,TK)/(AL, CK)

1.32 ± 0.06

P < 0.0011

Glomerular dis-tance (GD, HM)

161 ± 11

239±8P < 0.001*

214 ± 9

258±8P< 0.001*

(GD, TK)/(GD, CK)

1.33 ± 0.06

P<0.001t

Glomerular width (GW,JIM)

143 ± 2

140 ± 20.3 < P < 0.4*

163 ± 2

158 ± 20.05 <P< 0.1*

(GW, TK)/(GW, CK)

0.97 ± 0.009

P < 0.02f

Control kidneysn - 247 (104)

Transplant kidneysn = 398 (136)

Glomerular affer-ent art«nolar

width (AW, /IM)

9.8 ± 0.3

8.3 ± 0.2

p < 0.001*

(AW, TK)/(AW, CK)

0.88 ± 0.04

P < 0.05f

n *= number of measurements (number of slides analyzed).* Unpaired Z-test, comparing all measurements in control kidneys to all measurements in transplant kidneys.t Comparing mean ratio to 1.00.

AW, control kidney) for both outer and inner cortex.There was no difference between the mean ratiosfor outer or inner cortex; thus the ratios for outerand inner cortex were combined, averaged, and thismean ratio compared to 1.00, as shown in Table 1(bottom line). This analysis also shows that the AWof the transplant kidneys is significantly less thanthe AW of the control kidneys. Thus, the afferentarterioles of the transplant kidneys are stretchedand narrowed.

Changes in Renal FunctionThe serum creatinine increased in each of the

recipient dogs from an initial mean value of 1.0 ±0.1 to a mean value of 2.0 ± 0.4 mg/100 ml on thesixth posttransplant day. The range of individualvalues was +0.3 to +2.85 mg/100 ml. Because of therelatively large rise in serum creatinine in one dog,the standard error of the increase in serum creati-nine was relatively large, and thus paired ^-testingof the increase in serum creatinine did not show astatistically significant change (0.1 < P < 0.2). Onthe average, the transplanted kidney increased inweight by 140% (range, 110-210%). We assume thatthe increase in transplant kidney weight was duealmost entirely to the effects of renal allograft re-jection since the technique of transplantation itselfdoes not substantially affect kidney weight. Thiswas shown in a separate experiment in which weautotransplanted one of the dog's kidneys into theiliac fossa and then removed the contralateral kid-ney. We found that, 6 days post-autotransplanta-tion, the autotransplant kidney was only 10% heav-ier than its mate kidney removed at the time ofautotransplantation. In this dog, the serum creati-nine was 0.7 mg/dl at the time of autotransplanta-tion and 0.85 mg/dl by the sixth day post-autotrans-

plantation. The fact that the serum creatinine didnot change importantly indicates that absence ofswelling of the autotransplanted kidney was notdue to infarction of a major portion of that kidney,since this would have resulted in a much higher risein serum creatinine.

DiscussionThis study was undertaken to assess whether the

acute parenchymal swelling associated with renalallograft rejection results in stretching of the renalvasculature. Previous microangiographic studies oftransplant rejection have not evaluated this ques-tion (Almgard et al., 1966, 1967; Clark et al., 1977;Gardner et al., 1968). We compared the length andwidth of glomerular afferent arterioles (AL and AW,respectively) in normal dog kidneys to those of theircontralateral mate kidneys after they had beenallowed to undergo 6 days of unmodified rejection.We chose to measure the length and width of theglomerular afferent arteriole because this segmentof the vasculature is easily defined and because theglomerular afferent arteriole is a principal resistorin the renal vascular circuit (Abe et al., 1970). Wealso measured, in both control and transplant kid-neys, the shortest distance between the glomerulusand its feeding artery (GD) and the width of theglomerulus (GW).

We found that, on the average, AL was aboutone-third longer and, at its midpoint, about one-eighth narrower in the swollen rejecting kidneysthan in the normal kidneys. Furthermore, we deter-mined that the increase in AL was due almostentirely to stretching (i.e., an increase in truelength) of the glomerular afferent arteriole. Thisdetermination rests on the observation that, on theaverage, most glomerular afferent arterioles are

by guest on May 16, 2018

http://circres.ahajournals.org/D

ownloaded from

222 CIRCULATION RESEARCH VOL. 44, No. 2, FEBRUARY 1979

nearly straight as shown by the fact that mean GDapproximates mean AL. Thus, as discussed in detailin Results, uncoiling of the glomerular afferent ar-teriole (which results in an increase in length of theglomerular afferent arteriole visible to the camerabut does not stretch the glomerular afferent arter-iole) could not contribute importantly to the mea-sured increase in AL.

To the extent that uncoiling of the glomerularafferent arteriole occurred in the transplant kidney,the present study overestimated the degree ofstretching. It seems unlikely, however, that theestimated increase in transplant kidney AL is anoverestimate of the degree of afferent arteriolarstretching which actually occurred because the ef-fect of uncoiling on the measurement of AL wasprobably more than offset by overestimation ofcontrol kidney AL. An overestimation of controlkidney AL occurred because glomeruli with shortafferent arterioles are underrepresented in the mea-surement of AL in the control kidneys, for reasonsdiscussed in Results. Thus, since control kidney ALis overestimated, this resulted in a falsely low valuefor the ratio (AL, transplant kidney/AL, controlkidney), and this ratio was used as the measure ofthe degree of transplant kidney afferent arteriolestretching.

We also found that the transplant kidney afferentarterioles were significantly narrowed compared tothose of the control kidney. The finding of thehighly significant difference in AW between controland transplant kidneys is especially remarkable be-cause, of all the measurements, AW was measuredwith the least accuracy. This was unavoidable be-cause AW measurements involve relatively shortdistances and thus the blurred margins of the pro-jected image, even with optimum focusing, were asubstantial part of the total length measured. De-spite this source of variability in the measurement,a highly significant difference between transplantand control kidney AW was clearly demonstrated.

We also found that glomeruli were slightly re-duced in size in the transplanted kidney, comparedto the control kidney. It is unlikely that this findingis due to lesser filling of the transplant kidneyglomeruli since only those glomeruli which ap-peared to be completely filled from side to side (bydemonstrating smoothly rounded margins of theglomerulus) were chosen for measurement. It alsoseems unlikely that the smaller size of the glomeruliin the transplant kidneys is due to glomerularcompression since we have found that intrarenalpressure, as assessed by renal subcapsular pressuremeasurement (Hebert and Arbus (1971) and wedgerenal venous pressure measurement, is normal ornear normal (unpublished observations). Thus, themost likely reason for the smaller glomeruli in thetransplant kidney is the presence of lower glomer-ular capillary hydrostatic pressure. This could bethe result of an increase in afferent arteriolar re-sistance because of afferent arteriolar stretchingand narrowing.

The present study cannot assess the extent to

which afferent arteriolar stretching contributes tothe impaired renal function seen in swollen renalallografts since only the afferent arteriole was ex-amined and its diameter measured only at the mid-point. However, if the true average degree of affer-ent arteriolar stretching and narrowing were 1.32and 0.88, respectively, resistance to flow throughthe afferent arteriole would approximately double.*Clearly such a change in afferent arteriolar resist-ance could affect renal function substantially, re-gardless of adjustments elsewhere in the renal vas-cular bed.

In summary, as the renal cortex swells in acuteallograft rejection, glomeruli are pulled away fromtheir feeding arteries, their afferent arterioles be-come stretched and narrowed, and the glomerulibecome smaller. These anatomic findings indicatethat resistance to flow through the afferent arterioleis increased, resulting in a fall in glomerular hydro-static pressure and, presumably, filtration rate.Thus acute, compliance-mediated expansion ofrenal parenchymal volume accounts, at least inpart, for the impaired function of the swollen renalallograft undergoing acute rejection.

AcknowledgmentsWe gratefully acknowledge the secretarial assistance of

Sherry Nagel.

ReferencesAbe Y, Dixon F, McNay JL: Dissociation between autoregulation

of renal blood flow and glomerular filtration rate. Am J Physiol219: 986-993, 1970

Abrams HL: Quantitative derivates of renal radiologic studies.Invest Radiol 7: 240-279, 1972

Almgard LE, Granberg PO, Lagergren C, Ljungqvist A: Arteri-ovenous anastomoses in the canine renal allograft. Nephron3: 295-308, 1966

Almgard LE, Granberg PO, Lagergren C, Ljungqvist A: Theintrarenal vascular reaction of the canine renal allograft toimmunodepressive treatment Nephron 4: 32-44, 1967

Clark RL, Mandel SR, Webster WP: Microvascular changes incanine renal allograft rejection: A correlative microangio-graphic and histologic study. Invest Radiol 12: 62-73, 1977

Fletcher EWL, Chir B, Lecky JW: The radiological size of renaltransplants. A retrospective study. Br J Radiol 42: 892-898,1969

Gardner LB, Guttmann RD, Merrill JP: Renal transplantationin the inbred rat. Transplantation 6: 411-418, 1968

Hebert LA, Arbus GS: Renal subcapsular pressure. A new intra-renal pressure measurement. Am J Physiol 220: 1129-1136,1971

Hebert LA, Stuart KA, Stemper JA: Whole kidneyvolume/pressure relationships. Kidney Int 7: 45-54, 1975

Hebert LA, Stuart KA, Stemper JA: Effect of renal decapsula-tion on renal function. Am J Physiol 229: 632-639, 1975

Hebert LA, Riley AL, Itskovitz HD: Effect of expansion of renalparenchymal volume in renal function. Nephron, 1978 (inpress)

Schneider EG, Lynch RE, Willis LR, Knox FG: Single-nephronfiltration rate in the dog. Am J Physiol 222: 667-673, 1972

•R/R, - (Uwhere Ft, = resistance, t, = length, and ro = radius of vascular segment allunder control conditions.

R, — resistance, U ~ length, and r, — radius of vascular segment, allunder experimental conditions.

However, if 1, = 1.32 I, and r, - 0.88 im then.

R.JR, = [(U/ro4)]/[(l-321J/(0 88 rj4] - 2.17.

by guest on May 16, 2018

http://circres.ahajournals.org/D

ownloaded from

C L Allhiser, J E Stefaniak, L A Herbert, A J LaPorta, A L Riley and D Sampsonrejection.

Stretching of glomerular afferent arterioles in the swollen renal allograft undergoing acute

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1979 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.44.2.2161979;44:216-222Circ Res. 

http://circres.ahajournals.org/content/44/2/216.citationWorld Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on May 16, 2018

http://circres.ahajournals.org/D

ownloaded from