a definition of proximal and distal tubular compliance · diameter & pressure ( /im/mmhg) 0.gf-...

A Definition of Proximal and Distal Tubular Compliance

PRACTICAL ANDTHEORETICALIMPLICATIONS

STANLEYCORTELL, F. JOHNGENNARI, MICHAEL DAVIDMAN, WILLIAM H. BOSSERT, andWILLIAM B. SCHWARTZwith the technical assistance of MELV7IN L. PONTE

From the Department of Medicine, Tufts University School of AMedicine, andthe Renal Service of the New England Medical Center Hospital,Boston, Massachusetts 02111

A B S T R A C T Micropuncture studies were carried outin the rat to evaluate the in situ distensibility character-istics of the proximal and distal tubules under a varietyof experimental conditions. In the first phase, we de-termined the response of tubular diameter (D) to changesin tubular pressure (P) induced by partially obstructingsingle tubules. The response observed uncder these con-ditions (i.e., when interstitial pressure is presumed to beconstant) has been defined as the comnpliancc of the tu-bule. Over the range of tubular pressures studied (10-35 mmHg for the proximal tubule, 5-25 mmHg for thedistal tubule) the compliance characteristics of the prox-imal and distal tubule were found to be markedly dif-ferent; the proximal tubular pressure-diameter relation-ship was linear, AD/AP =0.45 gm/mm Hg, whereasthe distal pressure-diameter relationshiip was curvilinear,AD/AP =

In the second phase we used the comlpliance data toconstruct a series of theoretical pressure-diameter curvesthat define the response of the tubule to increments ininterstitial as well as intratubular pressure. Thesecurves indicate that changes in distal diameter shouldprovide a sensitive index of a rise in interstitial pres-sure under conditions in which the transtubular pressuregradient is increased by a small amount, but that proxi-mal diameter should provide a more sensiti-e index ofchanges in interstitial pressure when the transtubularpressure gradient is increasedl by a large amount. In

Dr. Cortell is the recipient of a Career DevelopmentAward (HE-35,834) National Heart and Lung Institute,National Institutes of Health; Dr. Gennari is the recipientof a Career Development Award (HE-38,908) NationalHeart and Lung Institute, National Institutes of Health;and Dr. Davidman is the recipient of a Fellowship fromthe Medical Research Council of Canada.

Received for putblicationt 9 Janitar 107?3 anid ill revisedformii. 17 April 1973.

subsequent experiments in which furosemide was ad-ministered, we observed that the pressure-diameter re-lationships for both the proximal and distal tubule wereindistinguishable from the compliance curves, a findingconsistent with the interpretation that interstitial pres-sure was not appreciably changed from control. By con-trast, when mannitol was administered, both proximaland distal tubular pressure-diameter relationships weresignificantly altered in a fashion consistent with a largeincrease in interstitial pressure. Neither with furosemidenor mannitol administration did it appear likely thatsignificant changes in tubular compliance could accountfor the observed behavior of the tubule.

Finally, we propose that a knowledge of tubular com-pliance will be useful in exploring the interrelationshipsbetween tubular and peritubular pressures, tubularanatomv, and transtubular ionic permeability. Recentstudies linking changes in the geometry of lateral inter-cellular spaces of the tubule to changes in passive ioInmovement suggest that an investigation of suclh anatomi-cal-ftunctional correlates should be productive.

INTRODUCTION

Recent evidence suggesting that intrarenal pressuresplay an important role in transtubular ion movement hascreated a renewed interest in the relationship betweentubular function and the physical properties and ana-tomical characterstics of the tubular epithelium (1, 2).Although the response of renal tubular structures tochanges in pressure has become a matter of considerableinterest, little is known about the in situ response oftubular diameter to increases in tubular pressure. In thefirst phase of the present study we have examined therelationship between tubular diameter and pressure overa wide range of increases in tubular pressure indtuced

Thle Joturnlal of Clinical Investigation Volume 52 September 1973 2330-23.392.330

by partially blocking single tubules. This relationship,which we have defined as tubular compliance, was foundto be notably different for the proximal and distaltubules.'

In the second phase of the study, we have used thesecompliance curves to construct a series of theoreticalpressure-diameter relationships defining the anticipatedbehavior of tubular diameter if interstitial as well asintratubular pressure were increased. These theoreticalcurves indicate that changes in distal diameter shouldprovide a sensitive indicator of increases in interstitialpressure under conditions in which the transtubular pres-sure gradient is increased by a small amount, but thatproximal diameter should be more sensitive to increasesin interstitial pressure under circumstances in which thetranstubular pressure gradient is large. Subsequent ob-servations during mannitol and furosemide diuresis,when analyzed within the framework of these theoreticalcurves, were consistent with the interpretation that in-terstitial pressure did not increase appreciably duringfurosemide diuresis but that it increased markedly dur-ing mannitol diuresis. The data also strongly suggestedthat drug-induced changes in compliance could notreasonably be invoked to account for the observed pres-sure-diameter relationships.

METHODSAll studies were performed on male Charles River rats(200-300 g) (Charles River Breeding Laboratories, Inc.,Wilmington, Mass.) fed a standard rat diet and allowedwater ad lib. After being anesthetized with Inactin 80-90mg/kg intraperitoneally, the animal was prepared for micro-puncture as described previously (3), except that the ureterwas catheterized with a PE50 rather than a PE10 catheter.

Upon completion of surgical preparation, all animals wererequired to meet the following criteria, or the experimentwas discontinued: (a) Blood pressure greater than 95 mmHg, (b) body temperature between 36-380C, (c) urinefree of gross blood, (d) transit time (by method of Stein-hausen [4]) less than 13 s with dye flowing evenly to allareas of the kidney and with no retention of dye in thetubules.

If the above conditions were met, 25 ,uCi of [14C] inulinwas given as a bolus, and [14C] inulin was added to thesustaining infusion in sufficient concentration to deliver 25,uCi/h. After a 30 min equilibration period, the experimentalprotocols described below were begun.

Experimental protocolsThe effect of increased intratubular pressure upon tubular

diameter was studied in the antidiuretic state in one seriesof experiments ("partial blockade studies") and in twohigh flow states in a second series of experiments ("diuretic

1 As we shall point out later, because the proximal anddistal tubules were necessarily studied over a differentrange of pressures, the difference in compliance observedfor these two segments of the nephron does not necessarilyimply a difference in their inherent distensibilities.

studies"). The two experimental protocols are described indetail below.

Partial blockade studies. In these experiments, increasesin intratubular pressure were induced in single proximal or

distal tubules by means of a castor oil block. The infusionbegun during surgical preparation (5% mannitol in 0.5%saline) was discontinued when surgery was completed, andan infusion of 0.5%o saline was given at 1.7 cm3/h for theremainder of the experiment.

Single tubules were studied in the following manner:Resting intratubular pressure was measured and the pres-sure-measuring pipette was left in place. Lissamine greendye was then infused into the tubule through this pipetteand photographs were taken (the details of the techniqueswill be described later.). After the photographs, the tubularpressure was remeasured. A pipette filled with castor oilwas then placed at a more distal site in the same tubuleand tubular flow was partially blocked by the insertion ofa long oil block. If tubular pressure after partial blockadeincreased by 5 mmHg or more in the proximal tubule or by2 mmHg or more in the distal tubule, the same sequenceof pressure, picture, and pressure measurement was thenrepeated.

Data from proximal tubules were not accepted unless thepressure after picture-taking was within 5 mmHg of thepressure measured before pictures were taken; data fromdistal tubules were not accepted unless the tubular pres-sures before and after picture taking were within 2 mmHgof each other. These different criteria for pressure changesin. the proximal and distal tubule were established on thebasis of preliminary studies in which it was noted thatduring partial blockade, tubular pressure was less stable inthe proximal tubule than in the distal tubule. The mean

difference between the pressures before and after pictureswas 0.25 mmHg for the proximal tubules (in 40% therepeat pressure was lower, in 55% the repeat pressure washigher, and in 5%o there was no change), and - 0.18 mm

Hg for the distal tubule (56% lower, 38% higher, 6% nochange).

Diuretic stuidies. Changes in intratubular pressure wereinduced by increasing tubular flow by infusion of eithermannitol or furosemide. The experiments were divided intothree periods: control, low-dose diuresis, and high-dosediuresis. In each period photographs were taken of thekidney surface for diameter measurements. Intratubularpressure measurements were made in at least one proximaland one distal tubule both before and after picture takingand in a total of not less than three proximal and threedistal tubules. In the control period, an infusion of 5%mannitol in 0.5% saline (begun during neck surgery) wasmaintained at 1.7 cm3/h. No notable diuresis was producedby this amount of mannitol as evidenced by a mean controlU/P inulin ratio of 378 and a mean urine flow rate of 9Atl/min. The two experimental periods for mannitol andfurosemide are described below.

Mannitol. In the low-dose diuretic period, 0.8 cm3 of10%o mannitol was given as a bolus, and a solution of 10%mannitol in 0.5% saline was then infused at 9 cm3/h. Inthe high-dose diuretic period 1 cm3 of 10%o mannitol wasgiven as a bolus, and the 10%o mannitol solution was theninfused at 18 cm3/h. In order to attain a relatively constanturine flow in each experimental period, a 30 min equilibra-tion period was allowed before pressure and diameter mea-surements were made.

Fuirosemide. In the low-dose diuretic period. 5 mg/kgof furosemide was administered as a bolus and the drug

Proximal and Distal Tubular Compliance 2331

DIAMETER& PRESSURE

( /im/mm H g) 0.Gf-

40

30

DIAMETER(am )

20

IC

S

-0. ocio3"s , 0 6'

15 20 25 30 35 40BLDCKED PRESSURE(mmHg)

lometer s 45 x pss +4 6

5 10 15 20 25 30 35 40PRESSURE(mmHg)

FIGURE 1 Proximal tubular compliance. In the upper panel,the change in diameter per mm Hg change in tubularpressure (AD/AP) is plotted against the absolute tubularpressure during partial blockade. In the lower panel, theabsolute diameter is plotted against tubular pressure duringcontrol and partial blockade; thus each tubule is repre-sented by two points. The lines drawn through the datapoints in both panels were calculated by the method of leastsquares.

then continuously infused in 0.7% saline at a rate of 5 mg/kg/h. In the high-dose period 25 mg/kg of furosemide wasgiven as a bolus, and then continuously infused in 0.7%saline at a rate of 50 mg/kg/h. In 1oth experimenitalperiods, a 20 min equilibration period was allowed beforepressure and diameter measurements were made.

During the two diuretic periods (in both the maninitoland furosemide experiments), urinary losses of fluid andelectrolyte were quantitatively replaced in order to minimizechanges in extracellular fluid volume. To approximate thesodium content of urinary losses indluced by each agent,0.5% saline was used for replacement in the mannitol ex-periments and 0.7%s salinie in the furosemide experimenlts.Data from an animal were accepted only if the fluid halance(total infusion minus urine flow for the two kidneys) 2 forthe entire 4-5 h experiment was 0-6 cm3 positive. Thisslightly positive fluid balance was allowed to account f orinsensible losses. The mean cumulative fluid balance was+4 cm3 for the furosemide experiments and +3 cm3 for themannitol experiments.

Techniques and analytical methodsProximal and distal tubular pressures were measured by

a modification of the technique of Gottschalk and Mylle (5)

'Urine flow of the two kidneys was estimated by doublin-measured single-kidney flow. To examine whether thisextrapolation was justified, urine flow rate, U/P inulin ratio,and GFR were measured for both the micropunctured kid-ney and the control kidney in four rats (two with mannitoland two with furosemide); no difference between the twokidneys was observed.

described previously f roml our laboratory (3). ['4C1 Inulinactivity in plasma and urine was measured as previouslydescribed (6).

7 lubular diameter mcasturemitents. Photographs of thekidniiey surface were taken on higb-speed Ektacbrome filmat X 40 magnificationi utilizing one ocular of a Leitz stereomlicroscope (E. Leitz, Inc., Rockleigh, N. J.). The fieldwas illuminated with a sy-nchronized Zeiss Ukatron 60Flash attachment (Carl Zeiss, Inc., Newv York). The photo-graphs for diameter measurement X-ere taken after intra-tubular lissamine green inj ectioIn in the partial blockadeexperiments, and after intravenous lissamine green injectionin the diuretic experiments.

Iiltratubular methlod (partial blockade stuidics). A 17%lissamine green solution was infused into individual tubulesby increasing the pressure in the pipette above the intra-tul)ular pressure until dye began to flow into the tubule. Tominimize any intratubular pressure chainges that might occurwith lissamine green inljection, the pipette pressure wasthen rapidly low-ered an(l the pictures were taken at a timewhen the pilette pressure was approximately equal to theintratubular pressures. Photographs w,ere Inot used fordiameter measurements unless the pressure in the pipetteat the time of picture-taking was within 5 mmHg of theintratubular pressure. In actuality, the mean pipette pres-sure was 0.9 mm Hg above the initratubular pressure forthe proximal tubule and 0.5 mmHg above the intratubularpressure for the distal tubule.

Intravenous method (diuretic sttudies). After the intra-venous injection of 0.05 cm3 of a 10%o lissamine greensolutioni, photographs were taken of dye-filled proximaland distal tubules. Siince suirface distal tubules are much lessnumerous than proximal, an attempt was made to photo-graph several different areas of the kidney surface aftereach injection. If the same distal tubule was photographedmore than onice, it was only measured in the first frame.

Tec1hniique of measurc)ncut. The photographs were pro-jected on a white cardboard screen witlh the projector kepta standard distance from the screen. Each roll of filmincluded a photograph of a stage micrometer taken at thesame magnification as the photographs of the kidney sur-face. All tubular diameters were arbitrarily measured asthe width of the intratubular green-dye column. The di-ameter measurements w-ere carried out by a team of twoobservers who had Ino kniowledge of which experiment orwhiclh experimental period they were measuring. The di-ameters were measured by one observer, Usilng calipers;the second observer recordled the caliper width. A 50 Iumsegment of the stage micrometer was also measured andrecorded for conversion of the diameter measurements tomicrometers. O(ne-quarter of all the diameters were re-measured by a team of two observers brought in fromoutside the laboratory. Their measurements were comparedwith the first group's measurements and no significant dif-ferences were noted.

Proximal tulules wvere measured in the followinig manner:In the intravenous injection series, each clearly-focused.dye-filled tubular loop w-as measured if the borders vereroughly parallel for at least 1 diam width. In the intratubu-lar injection series, a measurement was made in each seg-ment of the dye-filled loop if its borders were roughlyparallel. Loop segments containing a pipette were not mea-sured.

Distal tulelblcs were measured in the following manner: Inthe intravenous in;jection series, the widths of dye-filledtubules were measured in the same maniner as were the

2332 Cortell, Gennari, Davidman, Bossert, and Schwartz

l .0

. A

proximal tubules except that in those tubules with irregularborders, measurements were made both at the narrowest 5Cand widest places where the borders were roughly parallel;these two measurements were averaged. In the intratubular 4.cinjection series, each loop was measured in two to threeplaces, including the narrowest and the widest areas. Nomeasurements were made within 1 diam width of the ADIAMETER30pipettes. A PRESSUREpe m /mm Hg)

2.cRESULTS

Effect of partial blockade on pressure anddiameter in individual tubulesProximal tubule. The effect of increased tubular pres-

sure upon tubular diameter was studied by means of par-tial blockade in individual proximal tubules in 17 rats.The mean tubular pressure was 13.1±0.3 mmHg beforeblockade. In 30 out of 34 successfully blocked tubules,intratubular pressure increased by 5 mmHg or more.For each tubule fulfilling this criterion for increasedpressure (see Methods), the change in tubular diameterper mmHg change in tubular pressure (AD/AP) wasplotted against the absolute tubular pressure duringpartial blockade (upper portion of Fig. 1).' The slopeof the calculated regression line drawn through thesedata points was not significantly4 different from zero,indicating that the relationship between tubular diam-eter and pressure (lower portion of Fig. 1) was linearover the range of pressures studied. The mean value forAD/AP was 0.51±0.03 Am/mmHg (Fig. 3). The con-trol and "partial blockade" values for tubular diameterare plotted against tubular pressure in an unpaired fash-ion in the lower portion of Fig. 1. The slope of the cal-culated linear regression for these data, 0.45 Am/mmHg, was significantly greater than zero.

Distal tubule. The effect of increased distal tubularpressure upon tubular diameter was studied by meansof partial blockade in individual distal tubules in 16 rats.The mean tubular pressure was 6.7±0.3 mmHg beforeblockade. In all 23 tubules successfully blocked, intra-tubular pressure increased by the required 2 or moremmHg (see Methods). For each tubule studied, thechange in tubular diameter per mmHg change in tu-bular pressure (AD/AP) was plotted against the absolutepressure during partial blockade (upper portion ofFig. 2). As can be seen from inspection of this figure,in contrast to the behavior of the proximal tubule,AD/AP for the distal tubule decreased as tubular pres-sure was increased by partial blockade, indicating a non-linear relationship between tubular pressure and di-ameter.

' In one tubule in which intratubular pressure increasedby 5 mmHg, a negative value was obtained for AD/AP;this value was found to be an outlier variant (P <0.01)and was discarded (7).

'Throughout this paper the term "significant" will indi-cate a P value of less than 0.01.

0 5x pPmss o 21

1.0

30

DIAMETER(,am) 2C

lC

5 10 15 20BLOCKED PRESSURE(mmnHg)

25

Dia/er 565- e00*72'pr.1.036.50072

5 10 15

PRESSURE(mmHg)20 25

FIGURE 2 Distal tubular compliance. In the upper panel,the change in diameter per mmHg change in tubular pres-sure (AD/AP) is plotted against the absolute tubular pres-sure during partial blockade. In the lower panel, theabsolute diameter is plotted against tubular pressure duringcontrol and partial blockade; thus each tubule is repre-sented by two points. The curves drawn through the datapoints in both panels were calculated by means of a com-puter-based nonlinear least-squares search.

In the lower portion of Fig. 2, both the control and"partial blockade" values for tubular diameter areplotted against tubular pressure. From this representa-tion of the data, it can be seen that small increases inintratubular pressure above control were associated withlarge changes in tubular diameter and that progressivelylarger changes in pressure were associated with eversmaller additional increments in tubular diameter. Onthe basis of this pattern of diameter change, we havechosen an exponential function defining an asymptoticregression to represent the relationship between distaltubular diameter and pressure. It should be pointed out,however, that the shape of the resulting curve will beclosely similar whether an exponential function or aquadratic or higher-power polynomial function is usedto represent the data. In the equation for the exponentialfunction, D = C - 1/A X eAxP+B, the values for A, B,and C were calculated for our data points using a com-puter-based nonlinear least-squares search. This expo-


0

0 00 0

* .0

0.

0.

& Diameter& Pressure

(Ckm /mmHg) 0.

0.

-0.

P<O.O0 NS P<O.QI

FIGURE 3 Changes in proximal tubular diameter per mm

Hg increase in tubular pressure ( AD/AP) durinig partialblockade of single neplhrons (left-hand panel), maninitol di-uresis (center panel), and furosemide diuresis (riglht-hand(panel). Each point in the figure represents the change intubular diameter divided by the corresponditng change in

tubular pressure for each experimental observation in wlhichtubular pressure increased by 5 or more inm Hg. Thedashed line in each paniel indicates the mean value foreach group of observations. Note that the mean values forAD/A,P in the partial blockade and furosemide studies are

significantly greater than zero, whereas the mean value inthe mannitol study is not significantly differenit fromi zero.

nential function provided the basis for using the func-tion, AD/AP = exP+B, to describe the relatioinship be-tween AD/AP and the blocked pressure in the upper

portion of the figure. The values for and B in thelatter equation were also calculated froml our data usinigthe nonlinear least-squares search. The resulting equa-

tion, AD/AP = e--00'5o2 6 accouinted for 71%to of thetotal variation in AD/AP.

Effect of mannitol-induced diuresis on tubularpressure and diameterProximiial tutbutle. Proximal tubular pressure increased

from a mean value of 12.1 mmHg in the control periodto a mean value of 19.4 mmHg in the low-dose periodand rose further to 25.4 mmHg in the high-dose pe-

riod. In 15 determinations of proximal tubular di-ameter and pressure in 9 rats, proximal tubular pres-sure increased by 5 or more mm Hg. For these 15

determinations, as in the case of the proximal tubule in

the partial blockade studies, AD/AP did not vary signifi-cantly as a function of tubular pressure. The values forAD/AP during mannitol diuresis are shown in the cen-

ter panel of Fig. 3. The mean value, 0.10 Mm/mmHg,

was significantly less than the mean for AD/AP ob-served for the proximal tubule in the partial blockadestudies (left-hand panel of Fig. 3), a finding consistentwith the interpretation that interstitial pressure in-creased during mannitol diuresis.

Distal tubulc. Distal tubular pressure increased froma mean value of 6.8 mmHg in the control periodto a nmeani valuie of 14.8 mm Hg in the low-doseperiod and to 20.6 mmHg in the high-dose period.In 18 determinations of distal tubular pressure anddiameter in 9 rats, distal tubular pressure increasedby 2 or more mmHg above the control value. The meanvalue for AD/AP during mannitol diuresis, 0.99±0.09Am/mm Hg, was significantly less than the mean valuefor the distal tubule during partial blockade, 1.85±0.2uIm/nim Hg. In the left-hand panel of Fig. 4, AD/AP isplotted against the absolute pressure during mannitoldiuresis. The curve defining this relationship was, whentested by covariance analysis, significantly differentfrom the curve defining the relationship during partialblockade (dashed line) ; this difference was present re-gardless of wlhetlher a linear, quadratic, or exponentialfunction was used to define the partial blockade curve.5This difference in response is consistent with the inter-pretation that interstitial pressure increased duringmannitol diuresis. It should be noted that distal diameterduring nmannitol diuresis differed most strikingly fromthat observed in the partial blockade studies when tubu-lar pressure w-as increased by a small or moderateamount. At levels of tubular pressure above 20 mmHg,AD/AP in the two studies converged.

Effect of furosemide-induced diuresis on tubularpressure and diameterI'iroxiiiial tuibule. Proximal tubular pressure in-

creased from a mean value of 13.3 mm Hg in thecontrol period to a mean value of 18.5 mmHg inthe low-dose period but did not increase further inthe high-dose period. In 10 determinations of proximal

'As noted earlier, on the basis of the pattern of diameterchange an exponential function was chosen to represent therelationship between distal tubular diameter and pressure inithe partial blockade studies. However, if instead one utilizesa quadratic or cubic function to describe this relationship,the same statistical difference between mannitol and thepartial blockade studies is noted. This conclusion wasarrived at on the basis of the following analysis. For boththe quadratic and the cubic functions, the appropriate de-rivative functions were obtained and fitted to the data re-lating AD/AP to tubular pressure (left-hand panel of Fig.4) by a least-squares search. The resultant relationship,represented by a linear equation when the diameter-pressurecurve was described by a quadratic function and by a quad-ratic function when the diameter-pressure relationship wasdescribed by a cubic function, was then compared with themannitol data by means of a covariance analysis.


PARTIAL MANNITOL FUROSEMIDEBLOCKADE

.8

*

11 00*

2 o 0 -

I~~~~~~.2-

3.0\\ AoW101Blockade

\ A -O03 x,P,ss. */.52\ AP

*. \

Mannitol A* *\\* N

.

10 20 30

P-0

1.0

- Padrtial BOkode\ 'e

\* AD\

0

Furosemide \

N

\.\

*

W.+ /63

40 15 20 25

DIURETIC PRESSURE(mmH)

FIGURE 4 Changes in distal tubular diameter per mmHg increase in tubular pressure (ADIAP) during mannitol diuresis (left-hand panel) and during furosemide diuresis (right-handpanel). In each panel AD/AP is plotted against the absolute pressure during diuresis for eachexperimental observation in which tubular pressure increased by 2 or more mmHg. The solidlines drawn through the data points in each panel were calculated by means of a least-squaressearch. The dashed line in each panel represents the relationship between AD/AP and absolutepressure during the partial blockade experiments. See text for an explanation of the formof equation used to define the relationship during furosemide diuresis.

tubular diameter and pressure in 9 rats, proximal tubu-lar pressure increased by 5 or more mmHg. For these10 determinations, as in the case of the proximal tubulein the partial blockade experiments, ADIAP did notvary significantly as a function of tubular pressure. Thevalues for AD/AP during furosemide diuresis are shownin the right-hand panel of Fig. 3. The mean value, 0.39Am/mm Hg, was not significantly different from themean value for AD/AP observed for the proximal tubulein the partial blockade studies (left-hand panel of Fig.3), but was significantly greater than that observed dur-ing mannitol diuresis (center panel).

Distal tubule. Distal tubular pressure increased froma mean value of 7.6 mmHg in the control periodto a mean value of 14.3 mm Hg in the low-doseperiod, but as in the case of the proximal tubule,no further significant increase in pressure occurred inthe high-dose period. In 15 determinations of distaltubular pressure and diameter in 9 rats, distal tubularpressure increased by 2 or more mmHg above thecontrol value. For these 15 determinations the mean

value for AD/AP (1.39±0.15 Am/mm Hg) was notsignificantly different from the mean value for thedistal tubule in the partial blockade studies. In the right-hand panel of Fig. 4, AD/AP is plotted against theabsolute pressure during furosemide diuresis. The curve

defining this relationship, using either a linear or ex-

ponential function, was not significantly different from

the curve defining the relationship in the partial block-ade studies (dashed line). These findings indicate thatthe response of distal tubular diameter during furose-mide diuresis is indistinguishable from that observed inthe partial blockade studies; we have therefore usedan exponential function of the same type as describedearlier for the partial blockade experiments to definethe relationship during furosemide diuresis.

It should also be noted that the curve defining therelationship during furosemide diuresis was, when testedby covariance analysis, significantly different from thatobserved during mannitol diuresis (left-hand panel ofFig. 4).

Pressure gradient across the loop of Henle duringmannitol and furosemide diuresisFig. 5 presents the relationship between the pressure

gradient across the loop of Henle (proximal tubularpressure minus distal tubular pressure) and distal tubu-lar pressure before and during mannitol diuresis (left-hand panel), and before and during furosemide diuresis(right-hand panel). During mannitol diuresis the slopeof the calculated linear regression was not significantlydifferent from zero, indicating that the pressure gradientdid not decrease significantly as distal tubular pressure

increased. In contrast, the pressure gradient decreasedsignificantly after furosemide administration, falling


3.0-

2.01-

1.01-

i

:11

FUROSEMIDE

(Pprox - Pod,stJ - O06 x Pd/s, 5 7

20 30

(Pprox Pdist ) = - 0. 29 X Pd/sf 7' 8.2

20

DISTAL PRESSURE(mm Hg)

FIGURE 5 Pressure gradient across the loop of Henle during mannitol diuresis and furose-

mide diuresis. For each experimental period (control, low dose, high dose) in each rat, the

difference between the proximal and distal tubular pressure is plotted on the ordinate and the

distal tubular pressure is plotted on the abscissa. The lines (lrawn throughl the data poilntswere calculated bh the method of least squares.

from 7 to 2.5 mmHg when the distal tubular pressure

rose to 20 mmHg.

Urine flow rate and GFRIn the partial blockade experiments, mean urine flow

was 8±0.8 Al/min, mean U/P inulin ratio was 528427,and the mean GFR was 12.0±0.3 cm3/min/kg for two

kidneys. In the control period of the mannitol andfurosemide experiments, mean urine flow was 9±0.6Al/min, mean U/P inulin ratio was 378+31, and themean GFR was 13.7±0.6 cm3/min/kg for two kidneys.During mannitol diuresis, the mean urine flow rate afterequilibration was 178 Al/min in the low-dose period andincreased significantly to 497 il/min in the high-doseperiod. During furosemide diuresis, the mean urine flowrate was 325 ul/min in the low-dose period and in-creased significantly to 562 Al/min in the high-doseperiod. No significant changes in GFRwere noted after

the administration of either agent. It is noteworthy that

during mannitol diuresis, distal pressure increased sig-nificantly as urine flow increased, whereas during furo-

semide diuresis no increase in distal pressure was noted

after an increase in urine flow of similar magnitude.Expressed quantitatively, distal tubular pressure in-

creased by 0.032 mm Hg/Al increase in urine flow

(P < 0.01) during mannitol diuresis whereas no sig-

nificant correlation was noted between distal pressureand urine flow during furosemide diuresis.

DISCUSSIONThe present study has characterized the response of

tubular diameter when tubular pressure was varied

widely by partially blocking single tubules ini situt, that

is, under conditions of presumed constancy of interstitialpressure. We have defined the slope of the resultanitrelationship between tubular pressure and diameter

(AD/AP) as the comlpliance of the tubule. As illus-

trated in Figs. 1 and 2, the compliance characteristicsof the proximal and distal tubules were markedly dif-ferent. Proximal tubular diameter increased as a linear

function of the increase in tubular pressure, AD/ AP0.45. In contrast, the distal tubular pressure-diameterrelationship was curvilinear, tubular diameter virtually

doubling after a 5 mmHg increase in tubular pressure

and then increasing by ever smaller increnlents wvith

progressively larger increases in tubular pressure.'The compliance characteristics of the distal tubule

provide strong evidence that in the anitidiuretic state

interstitial pressur-e is less than distal tubular piressure.

' It should be noted that the difference in proximal anddistal tubular compliance after small increases in the trans-tubular pressure gradient (e.g. 5 mmHg) does not neces-

sarily imply a difference in inherent tubular distensibility.If, as seems highly probable, interstitial pressure is approxi-mately 6 mmHg in the antidiuretic state (see subsequentdiscussion in text), it follows that the proximal transtubularpressure gradient before blockade (6 mmHg) is already5 mm Hg greater than the distal transtubular pressuregradient. Thus, the range of transtubular pressures over

which the largest change in distal diameter occurred, i.e.,1-6 mmHg, cannot be examined for the proximal tubule.In the absence of such observations for the proximal tubuleone cannot decide w-hether or not the inherent distensibilitycharacteristics of the tw-o segments of the nephron are

di fferent.

2336 Cortell, Gennari, Davidman, Bossert, and Schwart;z

8.0 >-

*o

6.0k

40k

Qz~(I

Q.-0.

* .

20k

I0 30

MANNITOL

a -

i.e., less than 7 mmHg. Given that the distal tubuledilates so readily after a small increase in the trans-tubular pressure gradient, it seems extremely unlikelythat it would become rigid (i.e., not collapse) if thepressure outside were greater than the pressure inside.It is therefore difficult to reach any conclusion but thatresting interstitial pressure must be less than distalintratubular pressure. This interpretation is consistentwith recent estimates based upon subcapsular measure-ments of hydrostatic pressure that indicate a value forresting interstitial pressure of 6 mmHg (8).

The compliance characteristics of the distal tubulealso suggest that in certain settings distal diametershould be highly sensitve to changes in interstitial pres-sure. The predicted effect of a wide range of theoreticalincreases in interstitial pressure on the relationship be-tween distal tubular pressure and diameter is illustratedin the lower panel of Fig. 6. The solid line in thepanel was taken from the experimental measurement oftubular compliance7 and represents the relationship oftubular pressure to diameter when interstitial pressureremains constant. The dashed lines represent the pre-dicted tubular pressure-diameter relationships if inter-stitial pressure were increasing by a constant percent-age of the increase in tubular pressure (as indicated onthe right-hand side of the figure).

As would be anticipated from the curvilinear natureof the distal tubular compliance curve, the sensitivityof distal diameter to changes in interstitial pressure isgreatest with small changes in the transtubular pressuregradient and decreases markedly as the transtubularpressure gradient increases. Thus, it is apparent thatdistal diameter should be highly sensitive to changes ininterstitial pressure in any setting in which tubularpressure is increased by a small amount. The sensi-tivity of distal diameter in such a setting is illustrated bya consideration of points A and A' in Fig. 6. In thisexample, in which tubular pressure is assumed to in-crease by 5 mmHg, a detectable shift in tubular di-ameter (2 1m) away from the compliance curve (frompoint A to point A') would be predicted to occur ifinterstitial pressure were increased by as little as 1.5-2mmHg above control. With larger increments in tubu-lar pressure, the sensitivity of distal diameter to smallincreases in interstitial pressure progressively dimin-ishes, as is illustrated by points B and B'. In this latterexample, in which tubular pressure is increased by 10mmHg, a 2 Am shift in tubular diameter (from B toB') would be predicted to occur only after a 4-5 mmHg increase in interstitial pressure. With increments

'The distal tubular compliance curve was computed fromthe experimental measurements of AD and AP using acomputer-based nonlinear least-squares search. The resultingequation is: AD= 15.3 - 3.3 X ez8XAP+l S.

151

10

A DIAMETER()ULm)

5

0

A DIAMETEF( pm)

0o

AINTERSTITIAL

PRESSUREProximal Tubuile (%of ATubulor Pressure)0

_ 5 %

S - 70%- --~~ 0

5 10 15 20 25

50__ -370%

A -~~~~~~8/ ,- __- ..80%

/%/

'/

-- __90%/ / --

,/ - _-

.PL-IO05 10 15 20A TUBULAR PRESSURE(mmHg)

FIGURE 6 Predicted effect of a range of theoretical in-creases in interstitial pressure upon the proximal and distaltubular pressure-diameter relationships. The solid lines inboth panels are taken from the experimental measurementsof tubular compliance and represent the tubular pressure-diameter relationships when interstitial pressure remainsconstant. The dashed lines represent the predicted tubularpressure-diameter relationships if interstitial pressure wereincreasing by a constant percentage (as indicated on theright-hand side of the figure) of the increase in tubularpressure. Points A, A', B, B', and C in the lower panelare discussed in the text.

in tubular pressure of greater than 10 mmHg, it isevident that a detectable shift in distal diameter awayfrom the compliance curve would occur only if inter-stitial pressure were increased by greater than 50%of the increase in tubular pressure. By contrast, theproximal tubule (upper panel of Fig. 6) should beuseful in detecting relatively small increases in inter-stitial pressure even when tubular pressure is increasedby more than 10 mmHg. Because of the linear natureof proximal tubular compliance, a detectable shift indiameter (2 um) away from the compliance curve wouldbe predicted to occur whenever interstitial pressure isincreased by 4-5 mmHg.

If, on the other hand, interstitial pressure is in-creased by a large amount, obviously both proximaland distal diameter should readily detect the change.However, distal diameter should provide a more pre-cise estimate of the change when the increase in inter-stitial pressure is almost as large as the increase in


nictnl T,.hil,Ivl

TABLE I

Increase in Interstitial Pressure during MHannitol Diuresis asEstfimated from Changes in Tubular

Diamieter and Pressure

Estimated AA Tubular pressure A D interstitial pressure

Range Meatn Mean ±SEM Mean Range*

fnyn Hg ??I mmzn HgDistal tubule

3-8 6 6.6 ±+ 1.1 (6) 4 3-S12-15 13 10.9 -41.0 (6) 9 7-10

Proximal tubule10-14 12 0.6+(0.8 (8) 11 7-14

* Range of error obtained by using the meani diameter clanige ±2 SEMto estimate the increase in interstitial pressure.

tubular pressure (i.e., whein the net increase in thetranstubular pressure gradient is small). Thus, if tubu-lar pressure were increased by 15 mmin Hg ali(I inter-stitial pressure by 12 mmHg (point C in Fig. 6),distal tubular diameter would be as sensitive an indica-tor of changes in interstitial pressure as it was atpoint A' (when tubular pressure was increased by 5 mmHg and interstitial pressure by 2 mmHg). In eachinstance the estimate of the increase in interstitial pres-sure should be accurate to within +1 mmHg.

Estimate of the change in, intcrstitial pressute duringinfusion of furosentide or inannilitol. In subsequent ex-periments, the behavior of the proxinal and distal tubuleduring furosemide and mannitol diuresis was examinedin terms of the theoretical framework described above.During furosemide diuresis, AD/AP for both the proxi-mal and distal tubule (Figs. 3 and 4) was indistinguish-able from tubular compliance, suggesting that inter-stitial pressure did not increase. Since the increase intubular pressure was small (less than 7 immni Hg) inapproximately half the furosemide studies, we can as-sume (given the 95% confidence intervals around thedistal tubular diameter measurements), that if any in-crease in interstitial pressure occurred, it nmust havebeen less than 2 mmHg. In those furosemide studies inwhich the increase in tubular pressure Nas larger (7-15mmHg) our estimate is less precise, but analysis of theproximal tubular data in these studies suggest that ifany change in interstitial pressure occurred, it musthave been less than 5 mmHg.

It is obvious, of course, that the interpretation thatinterstitial pressure was unchanged during furosemidediuresis (an interpretation based on the agreement be-tween compliance and the furosemide pressure-diametercurves) is valid only if we assume that tubular compli-ance was not altered by the furosemide. This assump-tion seems reasonable since agreement between the

curves could have occurred in the face of a significantchange in interstitial pressure only if two highly un-likely conditioins were satisfied. First, furosemide wouldhave had to alter both proximal and distal tubular comi-pliance in a fashion so as to nearly exactly counterbal-ance the change in interstitial pressure; such behaviorseems extremely improbable in view of the very differ-ent structure and different compliances of these twoportions of the tubule. Second, the effect of furosemideupon tubular compliance would have had to have beenindependent of dosage since a fivefold increase in thequantity of furosemide administered had no notable in-fluence on the response of tubular diameter.

The pressure-diameter relationships during mannitoldiuresis were, in contrast to furosemide, consistent wvitha considerable change in interstitial pressure (Figs. 3and 4). As shown in Table I, in the studies with manni-tol in wlhich tubular pressure was increased by 3-8mmHg, distal tubular diameter changes were consistentwvith a concomiiitant increase in interstitial pressure of3-5 Imm Hg; in the studies in wliclh tubutlar p)ressurewas increase(l by 12-15 mmHg, distal diamiieter chanigeswvere conisistent wvith a concomitant increase in inlter-stitial pressure of 7-10 mIm Hg. It is notewvorthy thatin both the low-pressure and high-pressure experiments,the estimated increase in interstitial pressure was ap-proximately 70%, of the increase in tubular pressure.The data from the proximiial tubule, although much lessprecise, indicate a rise in initerstitial pressure of similarmagnitude (Table I).

With mannitol administration, as in the case of furo-semide, it seems highly improbable that the resultswere significantly influenced by an alteration in tubularcompliance. To account for the experimental findingsby a change in compliance, one would have to nmake theunlikely postulation that mannitol at each dosage levelaltered both proximal and distal tubular compliance insuch a way as to fortuitously mimic a uniform and pro-portionate increase in interstitial pressure.8

A comparison of estimated collecting-duct resistancein the two diuretic states also stronglv suggests thatinterstitial pressure increased more during mannitolthan during furosemide administration. Resistance toflow through the collecting ducts (R) can be calculatedfrom the equation, R = AP/F, wvhere AP is the pressuregradient across the collecting ducts and F is the flowthrough the collecting ducts. The pressure gradientacross the collecting ducts (AP) can be taken as equalto distal pressure since pelvic pressure is less than 1mmHg (3). Flow through the collecting ducts (F)

8 The conclusion that interstitial pressure is increasedduring mannitol diuresis is supported by the observationthat renal subcapsular pressure in the dog increases mark-edly during mannitol administration (8).


during the high-dose diuretic periods can be taken asequal to urine flow given that, at the observed highrates of urine flow (greater than 10% of GFR), col-lecting duct reabsorption is negligible. When the equa-tion is solved using these values, collecting-duct re-sistence is found to be almost 50% greater duringmannitol than during furosemide diuresis. This differ-ence in behavior indicates that collecting-duct diameterincreased less and (assuming no drug-induced changesin compliance) that interstitial pressure increased morewith mannitol than with furosemide administration.

A similar comparison of the estimated resistance toflow through the loop of Henle in the two diureticstates also provides support for the interpretation thatinterstitial pressure increased more with mannitol diure-sis than with furosemide diuresis. However, we are lessconfident of this interpretation than in the case of thecollecting duct, since our estimate of flow through theloop is much less precise. As shown in Fig. 5, thepressure gradient across the loop of Henle decreasedsignificantly as distal pressure increased during furo-semide diuresis but not during mannitol diuresis. If oneassumes that during administration of each drug proxi-mal tubular water reabsorption was decreased by anapproximately equal amount (9, 10) and that loopvater reabsorption was roughly the same, we can rea-

sonably conclude that loop resistance was significantlygreater (and interstitial pressure higher) during man-nitol than during furosemide administration. In fact,even if the increase in loop flow during furosemidediuresis was only half as large as during mannitoldiuresis, we would still be forced to the interpretationthat loop resistance and interstitial pressure weregreater during mannitol administration.

Finally, we should point out that the compliancecharacteristics of the tubules, through their influence oncellular configuration, may well play a significant rolein determining ionic permeability at various tubular andperitubular pressures. Boulpaep (1) has recently sug-gested that changes in the permeability of the proximaltubule to sodium and raffinose during saline diuresis aremediated through changes in the size and shape of thelateral intercellular spaces, a view supported by ana-tomical studies both in Necturus and in the rat (11, 12).Similar changes in the permeability characteristics ofthe proximal and distal tubule have been observed byother investigators in a variety of settings associatedwith increased tubular and peritubular pressure (13,14). Given this background, it seems reasonable tosuggest that a systematic examination of the interrela-tionships between tubular compliance, tubular anatomy,

and transepithelial ionic permeability should lead togreater insight into the control of net ion movement.

ACKNOWLEDGMENTThis work was supported in part by Grant HE759 fromthe National Heart and Lung Institute, National Institutesof Health.

REFERENCES1. Boulpaep, E. L. 1972. Permeability changes of the

proximal tubule of Necturus during saline loading.Am. J. Physiol. 222: 517.

2. Welling, L. W., and J. J. Grantham. 1972. Physicalproperties of isolated perfused renal tubules and tubularbasement membranes. J. Clin. Intvest. 51: 1063.

3. Cortell, S., M. Davidman, F. J. Gennari, and W. B.Schwartz. 1972. Catheter size as a determinant of out-flow resistance and intrarenal pressure. Am. J. Physiol.223: 910.

4. Steinhausen, M. 1963. Eine methode zur differenzierungproximaler und distaler tubuli der nierenrinde vonratten in vivo und ihre anwendung zur bestimmungtubularer str6mungsgeschwindigkeiten. Arch. Gesamte.Physiol. Menschen Tiere (Pflu(egers). 277: 23.

5. Gottschalk, C. W., and M. Mylle. 1956. Micropuncturestudy of pressures in proximal tubules and peritubularcapillaries of the rat kidney and their relation to ure-teral and renal venous pressures. Am. J. Physiol. 185:430.

6. Gennari, F. J., S. Cortell, and W. B. Schwartz. 1970.Loss of measured activity of inulin-C1' and a correctivetechnique. J. Appl. Physiol. 28: 105.

7. Snedecor, G. W., and W. G. Cochran. 1967. StatisticalMethods. Iowa State University Press, Ames, Iowa. 6thedition. 157.

8. Ott, C. E., L. G. Navar, and A. C. Guyton. 1971.Pressures in static and dynamic states from capsulesimplanted in the kidney. Am. J. Physiol. 221: 394.

9. Seely, J. F., and J. H. Dirks. 1969. Micropuncture studyof hypertonic mannitol diuresis in the proximal anddistal tubule of the dog kidney. J. Clin. Invest. 48: 2330.

10. Brenner, B. M., R. I. Keimowitz, F. S. Wright, andR. W. Berliner. 1969. An inhibitory effect of furosemideon sodium reabsorption by the proximal tubule of therat nephron. J. Clin. Invest. 48: 290.

11. Whittembury, G., F. A. Rawlins, M. Perez-Gonzales,and E. L. Boulpaep. 1972. Paracellular pathway for ionflow in kidney tubule. Abstracts from the 5th Inter-national Congress of Nephrology, Mexico City. 804.

12. Caulfield, J. B., and B. F. Trump. 1962. Correlation ofultrastructure with function in the rat kidney. Am. J.Pathol. 40: 199.

13. Lorentz, W. B., Jr., W. E. Lassiter, and C. W. Gott-schalk. 1972. Renal tubular permeability during in-creased intrarenal pressure. J. Clin. Invest. 51: 484.

14. Bank, N., W. E. Yarger, and H. S. Aynedjian. 1971. Amicroperfusion study of sucrose movement across therat proximal tubule during renal vein constriction. J.Clin. Invest. 50: 294.


a definition of proximal and distal tubular compliance · diameter & pressure ( /im/mmhg) 0.gf-...

Documents