Journal of Clinical InvestigationVol. 45, No. 1, 1966

Patterns of Nephron Perfusion in Acute and ChronicHydronephrosis *

WADI SUKI, GARABEDEKNOYAN,FLOYDC. RECTOR,JR., ANDDONALDWV. SELDIN t

(From the Department of Internal Medicine, The University of Texas Southwestern MedicalSchool, Dallas, Texas)

Partial urinary tract obstruction is known toproduce, in addition to a nonspecific depression inrenal function, certain characteristic derangements,especially profound polyuria (1-5); release of theobstruction may result in a transient salt-wastingstate (6, 7). Three hypotheses may be advancedto explain the altered function of the hydro-nephrotic kidney. It is possible that with the ele-vation of intrapelvic pressure a significant frac-tion of the nephron mass stops functioning, dueeither to mechanical damage or cessation of filtra-tion, whereas the remaining nephrons continue tofunction with decreased tubular flow. This de-creased rate of tubular flow could be due to re-duced filtration per nephron or increased proximalfractional reabsorption, or both. This hypothesiswould imply a reduced delivery of filtrate to theloop of Henle and the distal convoluted tubule andwould, therefore, result in a pattern of nephronunderperfusion (8). An alternative possibility isthat the initially elevated intrapelvic pressure re-sults in a reduction in nephron mass; the subse-quent return of intrapelvic pressure towards nor-mal might produce a compensatory increase in therate of tubular flow in the residual nephrons (9).The increased rate of tubular flow could resultfrom increased filtration per nephron or diminishedproximal fractional reabsorption, or both. Theincreased tubular flow per nephron increases the

* Submitted for publication August 19, 1965; acceptedOctober 13, 1965.

Supported in part by grants I-SOI-FR 5426 and 5TIHE-5469 from the National Institutes of Health, U. S.Public Health Service.

Presented in part at the National Meeting of theAmerican Society for Clinical Investigation on May 2,1965; a preliminary report appeared in J. clin. Invest.1965, 44, 1103.

t Address requests for reprints to Dr. Donald W.Seldin, Dept. of Internal Medicine, The University ofTexas Southwestern Medical School, Dallas, Texas75235.

delivery of filtrate to the distal nephron resultingin overperfusion of that portion. Finally, varyingdegrees of nephron damage might be superimposedon either of the two patterns of nephron perfusion.No studies are available to determine which ofthese theoretical considerations best fits the ob-served disorders of renal function.

This study was undertaken, therefore, to clarifythe pattern of renal function in hydronephrosis.The excretion of salt and water during water andhypotonic saline diuresis was examined in dogswith either acute or chronic unilateral hydrone-phrosis; the contralateral kidney served as an in-ternal control. The results from these two modelsof hydronephrosis were compared to those froman experimental model known to result in nephronunderperfusion, namely, renal arterial constriction(8).

Methods

The effects of acute and chronic hydronephrosis andof renal arterial constriction were studied in 14 femalemongrel dogs weighing 8 to 15 kg and fed commercialdiets. A total of 22 studies was performed during water

diuresis followed by hypotonic saline diuresis.Water diuresis was induced by the administration of

50 ml of water per kg body wt via a gastric tube andmaintained by the intravenous infusion of 2.5% solutionof dextrose in distilled water at a rate of 10 ml per min-ute. When the water diuresis was well established, two

or three urine and blood collections were made. Hy-potonic saline solution (0.45 to 0.58%) was then startedat a rate of 20 ml per minute, and 30 to 45 minutes latertwo or three more urine and blood collections were

made. After the termination of the urine collections dur-ing hypotonic saline diuresis, the maximal tubular trans-

port rate for p-aminohippurate (TmPAH) was measuredin four experiments during acute hydronephrosis andfour experiments during chronic hydronephrosis by theinfusion of loading and maintenance doses of PAH suffi-cient to produce a plasma level of 11 to 30 mg per 100 ml.This insured a load to Tm ratio of at least 2 to 3, even

in the presence of marked reductions in renal blood flow(10).

122

RENAL FUNCTION IN ACUTEAND CHRONICHYDRONEPHROSIS

Acute hydronephrosis was induced in seven dogs by theelevation to variable levels (60 to 80 cm H20) of a poly-ethylene tube inserted in the left ureter. The animalswere first anesthetized with 30 mg per kg pentobarbitalint-venously, and then through a midline abdominal in-cisiun each ureter was cannulated with a polyethylenetube reaching into the renal pelvis.

The renal artery was constricted in four anesthetizedanimals. Through a left flank incision, a silk tie wasplaced around the left renal artery, which was constrictedsufficiently to reduce blood pressure distal to the con-striction by 50%. The blood pressure was monitored bya small needle in the renal artery distal to the point ofconstriction, connected to a Statham strain gauge and aSanborn recorder. Both ureters were cannulated withpolyethylene tubing.

In six dogs a bladder-splitting operation was first per-formed to permit the collection of urine from each kidneyseparately (11). Nine to 12 days later a control studywas performed to ascertain whether glomerular filtrationrate (GFR), sodium excretion, and free water clearance(CH20) were equal on the two sides. Through a para-median abdominal incision the left ureter was then ex-posed and a silk tie placed around it, partially occludingit in the region of the ureteropelvic junction. In one dogthe induction of chronic hydronephrosis was precededby the cannulation of the left ureter and a study of acutehydronephrosis. These chronically hydronephrotic ani-mals were then studied 7 to 124 days after the constric-tion of the ureter.

The GFR was measured by the clearance of inulin.Inulin, PAH, sodium, potassium, and urinary soluteconcentrations were measured by previously publishedmethods (8). The data were analyzed by a nonparametricmedian test (12).

Results

Figure 1, constructed from our previously pub-lished data (8), depicts the effects of acute renalarterial -constriction. During water diuresis GFR,urinary flow, and urinary sodium concentration(UNa) were always lower, and total solute concen-tration (Uosm) always higher, on the constrictedthan on the control side. During hypotonic salinediuresis, two striking changes were seen: first, UNaon the control side rose sharply, but little or nochange was noted on the constricted side; second,UOsmon the constricted side, which was above thecontrol side during water diuresis, fell to a valuebelow that of the control side.

A typical experiment during acute hydronephro-sis is shown in Figure 2. The results of this studydisclose a pattern similar to that of renal arterialconstriction. During water diuresis GFR, urinaryflow, and UNa were lower, whereas Uosm washigher, on the hydronephrotic than on the control

0

FIG. 1. THE EFFECTS OF RENALARTERIAL CONSTRICTIONON WATERANDELECTROLYTEEXCRETIONDURINGWATERANDHYPOTONICSALINE DIURESIS. GFR= glomerular filtrationrate.

side. During hypotonic saline diuresis UNa onthe normal side rose with little or no change on thehydronephrotic side, and Uosm fell from a valueabove that of the normal side during water diure-sis to below the normal side during hypotonicsaline diuresis.

A representative experiment on an animal withchronic hydronephrosis is shown in Figure 3. Incontrast to acute hydronephrosis, UNa was mark-edly higher on the hydronephrotic than the con-trol side during both water and hypotonic salinediuresis. The failure of the U..m on the hydro-nephrotic side to fall from a value above that of thecontrol side during water diuresis to below thecontrol side during hypotonic saline diuresis alsois in marked contrast to the pattern seen in bothrenal arterial constriction and acute hydro-nephrosis.

The effects of all three experimental models onthe process of urinary dilution are shown in Fig-ure 4 and the results expressed as the ratio of theexperimental to the control side. Renal arterial

123

SUKI, EKNOYAN,RECTOR, AND SELDIN

Dog M

WATER HYPOTONIC SALINE

.O

0 F40 _

20A ~ :8--.---

0a 1§

FIG. 2. THE EFFECTS OF ACUTE HYDRONEPHROSISON

WATERAND ELECTROLYTE EXCRETION DURING WATERAND

HYPOTONICSALINE DIURESIS.

constriction and acute hydronephrosis yielded simi-lar results: the Uosm from the experimental sidewas invariably higher during water diuresis andlower during hypotonic saline diuresis than that ofthe control side. In contrast, during chronic hy-dronephrosis UOsm was higher on the hydrone-phrotic side in both water diuresis (6 of 11 stud-ies) and hypotonic saline diuresis (9 of 11 stud-ies). The difference in Uosm between acute andchronic hydronephrosis during hypotonic saline

Normal Kidney0-S-M

HydronephroticKidney

0-0-0

GFRmI/min

40

Dog 19

WATER I HYPOTONICSALINE I

20 _

10URINE FLOW

ml/min

201 _.

a00 ~~~~~~~~~~~~~~~~4

URINE SODIUMCONCENTRATION

mEq/L 2

15

TOTAL 10

URINARY SOLUTEmOsm/L 5

OC-O0o

FIG. 3. THE EFFECTS OF CHRONIC HYDRONEPHROSISON

WATERAND ELECTROLYTE EXCRETION DURING WATERAND

HYPOTONIC SALINE DIURESIS. This experiment was per-

formed 5 weeks after permanent ureteral constriction.

diuresis was highly significant with a p value of< 0.005.1 The effects of renal arterial constric-tion and acute hydronephrosis on free water clear-ance (CH2o) were also similar. During both waterand hypotonic saline diuresis the absolute CH2o

1 Henceforth, the p values reported refer to the differ-ence between acute and chronic hydronephrosis.

3

2

(Uosm) EXP(Uosm) CONT

RAC= RENALARTERYCONSTRICTION (4 EXPS)AH = ACUTEHYDRONEPHROSIS(7 EXPS)CH= CNRONICHYDRONEPHROSIS(I I EXPS)

(CH2O) EXP

(CH2O)CONT.8

.6

.4.

.2

FIG. 4. THE EFFECTS OF RENAL ARTERIAL CONSTRICTION ANDACUTEANDCHRONICHYDRONEPHROSISON URINARY SOL-

UTE CONCENTRATION(U.sm), ABSOLUTESOLUTE-FREE WATERCLEARANCE(CH2o), ANDFRACTIONALFREEWATERCLEARANCE

(CH2o X 100/GFR). The vertical bars represent the mean of the ratios of the experimental to the control sidestandard error.

124

L35 1~

25 -

5

Normal Kidney0-@

HydronephroticKidney

0-0-

GFRmVmin

URINE FLOWml/min

-~~~~~~

URINE SODIUMCONCENTRATION

mEq/L

TOTALURINARY SOLUTE

mOsm/L

200 1-

100p

D---*0... Name

e0000000."Owwgo-----O

;

0 -Noma

IIaI 0---O

RENAL FUNCTION IN ACUTEAND CHRONICHYDRONEPHROSIS

(Figure 4, middle panel) and the fractional excre-tion of free water [CHoG per 100 ml GFR (Figure4, panel on the right)] were always lower on theside with renal arterial constriction or acute hy-dronephrosis than on the control side. In chronichydronephrosis, by contrast, although the absoluteCH20 was lower on the experimental than on thecontrol side (Figure 4, middle panel), CHOper100 ml GFRwas greater on the hydronephroticthan on the control side (Figure 4, panel on theright) in 8 of 10 studies during water diuresis(p < 0.01) and 7 of 10 studies during hypotonicsaline diuresis (p.< 0.05).

The effects of renal arterial constriction andacute and chronic hydronephrosis on urinary so-dium are shown in Figure 5. Urinary sodiumconcentration was always lower on the side withrenal arterial constriction than on the control sideduring both water and hypotonic saline diuresis.In acute hydronephrosis a pattern similar to thatof renal arterial constriction was seen in all ex-periments. In contrast, in chronic hydronephrosisthe urinary sodium concentration was alwaysgreater on the hydronephrotic side than on the

control side (p < 0.05 during water diuresis and< 0.005 during hypotonic saline diuresis).

The absolute rate of sodium excretion (UNaV)and the fractional rate of sodium excretion [ (CNa/inulin clearance) X 100] were always lower on theside with renal arterial constriction than on thecontrol side (Figure 5). A similar effect on theabsolute and fractional rates of sodium excretionwas obtained in the experiments with acute hydro-nephrosis. Chronic hydronephrosis, on the otherhand, resulted in higher absolute rates of sodiumexcretion in 9 of 10 experiments during water diu-resis (p < 0.01) and 6 of 10 experiments duringhypotonic saline diuresis (p < 0.005); the frac-tional rate of sodium excretion was uniformlygreater on the side- with chronic hydronephrosisthan on the control side (p < 0.01 during waterdiuresis and < 0.005 during hypotonic salinediuresis).

The effect of acute and chronic hydronephrosison the relationship between nephron perfusion andtubular mass was also examined. The GFR onthe experimental, as compared with the control,side was reduced 24 to 42%o in renal arterial con-

WWATER RAC= RENALARTERYCONSTRICTION (4 EXPS)AH = ACUTEHYDRONEPHROSIS(7 EXPS)

HYPOTONICSALINE CH = CHRONICHYDRONEPHROSIS(II EXPS)

RAC AH CH RAC AH CH RAC AH CH

108

6

4

2

uIll¶Y -t,

.8 F(UNaV) EXP

(UNOV) CONT

FIG. 5. THE EFFECTS OF RENAL ARTERIAL CONSTRICTION AND ACUTEAND CHRONICHYDRONEPHROSISON URINARY SO-DIUM CONCENTRATION(UN), AND ABSOLUTE (UN.V) AND FRACTIONAL (CNx X 100/INULIN CLEARANCE) RATES OF SO-DIUM EXCRETION. The vertical bars represent the mean of the ratios of the experimental to the control side ± stand-ard error.

[UN,] EXP

[UN,] CONT

125

.1

.1

.

SUKI, EKNOYAN, RECTOR, AND SELDIN

striction, 14 to 77% in acute hydronephrosis, and5 to 92%o in chronic hydronephrosis. The TmPAHEon the experimental side was reduced below thatof the control side in both acute and chronic hy-dronephrosis. The relative effects of acute hydro-nephrosis and of chronic hydronephrosis on theGFRand TmPAHexpressed as a ratio are shownin Table I. It is apparent that in acute hydro-nephrosis the ratio of the GFR to TmPAHwasconsistently reduced on the hydronephrotic sideas compared to the control side; the ratio on theexperimental side was 9 to 37% lower than onthe control side, indicating that there was a dispro-portionate fall of GFR compared to TmPAH.These findings in acute hydronephrosis are similarto those of Thompson, Barrett, and Pitts in ani-mals where renal underperfusion was induced bythe inflation of a balloon in the aorta (13). Inchronic hydronephrosis a different pattern was evi-dent; the ratio of GFRto TmPAHwas 18 to 76%ohigher on the hydronephrotic side than on thecontrol side.

Discussion

The terms nephron overperfusion and nephronunderperfusion are usually taken to mean in-creased or decreased GFRper nephron. For pur-poses of the present study, however, where thediluting capacity of the kidney is used as an indexof perfusion, it is more useful to redefine theseterms in a -more restrictive sense as the amountof fluid delivered to the diluting segment, withoutspecifying the mechanism that might be responsiblefor alterations in delivery rate. This requiresthat a standard be available in terms of which de-livery rate to the diluting segment can be assessed;the contralateral control kidney serves this pur-pose. Thus redefined, underperfusion and over-perfusion mean, respectively, reduced or in-creased tubular flow through individual distalnephrons,2 as compared to the flow of the contra-lateral control kidney.

2 For purposes of this discussion the distal nephron isdefined as that portion of the nephron that is imperme-able to water during maximal suppression of antidiuretichormone (ADH). This presumably includes the as-cending limb of Henle's loop, the distal convoluted tu-bule, and the collecting duct. The proximal nephron in-cludes those segments proximal to the water-impermeablesegment and consists of the proximal convoluted tubule,pars recta, and the descending limb of Henle's loop.

Constriction of the renal artery in the presentexperiments reduced the glomerular filtration rateby 24 to 42%. Although it has been demonstratedthat glomerulotubular balance is maintained undersuch circumstances so that the percentage of fil-trate reabsorbed in the proximal tubule remainsconstant (14-16), the absolute delivery of filtrateto the distal nephron is sharply reduced. Con-striction of the renal artery, therefore, furnishesan excellent model to examine the pattern of uri-nary formation during underperfusion of the distalnephron.

Two consequences result from diminished de-livery of filtrate to the distal nephron. First, so-dium reabsorption is more complete and thereforethe formation of solute-free water per unit volumeof tubular fluid is enhanced. Second, back-diffu-sion of free water in the collecting duct per unitvolume of tubular fluid is augmented, thereby in-creasing free water loss. Which of these processeswill predominate during slowed flow depends prin-cipally on whether sodium reabsorption in the di-luting segment is already functioning near itslimiting concentration gradient. During waterdiuresis the concentration gradient of sodiumacross the distal nephron is near a maximal value.Slowed flow will augment sodium reabsorptionslightly; this will increase free water formationper unit volume only minimally, but will markedlylower UNa. The absolute volume of free waterformed is diminished due to the reduction of thevolume of fluid delivered to the diluting segment.An additional effect of slowed flow during waterdiuresis will be seen in the collecting duct wherethe loss of the same, or even reduced, absoluteamounts of free water from a markedly diminishedvolume of fluid will result in a sharp reductionin CH2Oper unit volume of tubular fluid and a risein Uosm.

During water diuresis, therefore, the side withrenal arterial constriction will display a reducedrate of CH20 because reduced delivery of filtrateto the diluting segment diminishes free water for-mation; a lower UNa, because the magnitude ofreabsorption during slowed flow, though small inabsolute amount, is proportionately greater thanthe rate of back-diffusion of free water; and a

higher Uosm because free water back-diffusionconcentrates nonsodium solutes, principally urea,

126

RENAL FUNCTION IN ACUTEAND CHRONICHYDRONEPHROSIS

even though the excretion of urea is sharply re-duced owing to a fall in GFR.

During hypotonic saline diuresis without renalarterial constriction (Figure 1, normal kidney)expansion of extracellular fluid volume results inincreased delivery of tubular fluid to the distalnephron; this is associated with a greater produc-tion of free water and rise in UNa with little changein Uosm. The constancy of Uosm while UNa risessharply is a result of the dilution of nonsodium uri-nary solutes by the increased excretion of sodium-containing urine.

During hypotonic saline diuresis the superimpo-sition of renal arterial constriction results in a fallin the urinary concentrations of both sodium andtotal solute (Figure 1, panel on right). The re-duced Uosm is the opposite effect from that ob-served when the renal artery is constricted duringwater diuresis. Since, under the condition of in-creased sodium delivery (hypotonic saline diure-sis), the diluting segment is operating above itslimiting concentration gradient, the greatly in-creased reabsorption of sodium as a result ofslowed flow (renal arterial constriction) will aug-ment the formation of solute-free water per unittubular volume. In consequence, UNa and U1,mfall. The fall in Uosm on the constricted side dur-ing the transition from water diuresis to hypotonicsaline diuresis (Figure 1) is attributable to dilu-

28cc

0

E

SNS

EuiUz

a:

3:LusCZ

24

20

tion of nonsodium urinary solutes by augmentedurinary flow.

These findings define the physiologic pattern ofdistal nephron underperfusion. During waterdiuresis acute reduction in GFR results in de-creased UNa (Figure 5) and increased Uo.m (Fig-ure 4), whereas in hypotonic saline diuresis re-duced GFRdiminishes both UNa and Uosm; CH2Oper 100 ml GFRis reduced during water diuresisand hypotonic saline diuresis (Figure 4). Sincethe findings in acute hydronephrosis are similarto those in renal arterial constriction, we concludethat acute hydronephrosis produces underperfusionof the distal nephron.

The pattern of renal function observed in chronichydronephrosis differed markedly from that re-sulting from renal arterial constriction and acutehydronephrosis. The chronically hydronephrotickidney compared with its contralateral control dis-played a higher UNa, a higher absolute and frac-tional UNaV (Figure 5), a higher Uo.m during wa-ter diuresis that did not fall during hypotonic sa-line diuresis'(Figure 4), and a lower CHIo (Fig-ure 4).

This pattern of renal function in chronic hydro-nephrosis could represent the superimposition oftubular damage on the distal underperfusion char-acteristic of acute hydronephrosis. Tubular dam-age, by impairing sodium transport in the diluting

* CONTROLO CHRONIC HYDRONEPHROSIS

0 0

16k

12-

4

00

00 dado*0

*oO

0 *

0a00 0

*:o 00

0 000 ~~~~0

*1 0

00 0Oo°0

4 8 12 16 20 24 28 32URINE FLOW ml/min/100ml GFR

36 40 44

FIG. 6. FRACTIONAL FREE WATER CLEARANCE AT VARYING FRACTIONALRATES OF URINARY FLOWIN CHRONICHYDRONEPHROSISAS COMPAREDTO NOR-MAL. Points represent values obtained during both water and hypotonicsaline diuresis.

127

0a

8p~

SUKI, EKNOYAN, RECTOR, AND SELDIN

TABLE I

The ratio of glomerular filtration rate (GFR) to maximaltubular transport rate for p-aminohippurate (TmPAH)

in acute and chronic hydronephrosis

GFR/TmPAH

Experimental Experi-model mental Control Exp./Cont.

Acute 6.3 8.3 0.8hydronephrosis 4.8 7.7 0.6

7.1 9.9 0.79.5 10.5 0.9

Mean ±SD 6.941.7 9.141.1 0.75*40.1

Chronic 10.7 8.3 1.3hydronephrosis 12.5 7.1 1.8

13.0 8.8 1.55.4 4.6 1.2

Mean -SD 10.4 43.0 7.2 ::1.6 1.45*40.2

*p <0.01.

segment, could result in the high UNa, high Uo.m,and low CH2o characteristic of chronic hydro-nephrosis.

To examine the functional integrity of the di-luting segment, we plotted Cia2o per 100 ml GFRagainst urinary flow .(V) per 100 ml GFR (Fig-ure 6). CH20 was plotted against V rather thansolute clearance (Co.m) because the rate of urinaryflow in water diuresis is a closer approximationof the rate of delivery of filtrate to the loop ofHenle than is Cosm.3 If free water formation were

significantly impaired, less CH20 should be foundon the hydronephrotic than on the normal sideat any given rate of urinary flow. This is, how-ever, not the case. It is apparent from Figure 6that, at comparable rates of delivery, the normaland chronically hydronephrotic kidneys generatethe same CH20. Therefore, there is no apparentdefect in CH20 in chronic hydronephrosis.

Since CH20 represents- a balance between freewater formation and free water back-diffusion, it

might be argued that a defect in formation couldbe masked by a commensurate reduction in back-diffusion. The shrinkage and scarring of themedulla so characteristic of chronic hydronephro-sis undoubtedly prevents the establishment of

3 During water diuresis the distal convoluted tubule andthe collecting duct are relatively impermeable to water.The rate of urinary flow, therefore, is the best index ofthe rate of delivery of tubular fluid to the distal nephron.Urinary flow, however, does not equal the rate of de-livery since some back-diffusion of water from the col-lecting duct is known to occur even in the absence ofADH (8, 11, 17, 18).

medullary hypertonicity and impairs diffusion ofwater out of the collecting duct. However, re-duction in free water loss would contribute equallyto urinary flow and CH20. Consequently, althougha reduction in free water back-diffusion would in-crease CH2o, it would not correct the low CH20 toa normal value when C1120 is compared to urinaryflow. Since the relations between CH20and urinaryflow in the normal and hydronephrotic kidney aresuperimposable, this would constitute evidenceagainst the hypothesis that the pattern of urinaryformation in chronic hydronephrosis is attributableto a combination of underperfusion of nephrons,tubular damage, and reduced free water back-dif-fusion. None of these arguments excludes the pos-sibility that some degree of tubular damage mightexist, and that free water back-diffusion might bereduced. Indeed, the profound destruction of themedulla almost certainly means that the normalback-diffusion of free water is drastically curtailed.Wemerely wish to emphasize that the combina-tion of tubular damage and reduced free waterback-diffusion cannot readily explain the strikingincrease in sodium excretion in the face of normalCH120 in the setting of distal nephron under-perfusion.

The changes in sodium and water excretion inchronic hydronephrosis can be more easily ex-plained on the basis of reduced nephron mass withoverperfusion of residual nephrons. It is not un-reasonable to assume that the GFRper nephron inchronic hydronephrosis is elevated. Although in-trapelvic pressure rises sharply with acute obstruc-tion of the ureter, it has been shown that with thepersistence of obstruction the intrapelvic pressurefalls gradually towards normal values (9). De-spite the fall in intrapelvic pressure some nephronsnever regain function and the functional renal massis reduced. When the intrapelvic pressure falls,however, the remaining nephrons of the hydro-nephrotic kidney may respond in a similar man-ner to those of a kidney damaged from any othercause by a compensatory increase in the rate offiltration per nephron (19).

Furthermore, the data relating the GFR to re-nal tubular mass (TmPAH) are consistent with thepresence of increased GFRper nephron in chronic,as distinct from acute, hydronephrosis (Table I).In renal underperfusion due to inflation of a bal-loon in the aorta a fall in maximal tubular trans-

128

RENAL FUNCTION IN ACUTEAND CHRONICHYDRONEPHROSIS

port of glucose (TmgiucO,,,) (13) has been re-ported. Acute elevation of ureteral pressure hasalso been reported to result in a reduction inTmglu,08e (10) and TmpAH [(10) and presentstudy]. In both experimental procedures the fallin GFR almost always exceeded the fall in thetubular maximum, thus suggesting a reduced fil-tration rate per nephron. In chronic hydronephro-sis, however, the fall of TmPAHwas always in ex-cess of the fall in GFR. The increased GFRperTmPAHmay be interpreted as an increased filtra-tion rate per nephron.

It is likely, therefore, that the combination of anelevated GFR per nephron and diminished frac-tional reabsorption of filtrate in the proximalnephron results in marked overperfusion of thedistal nephron. The larger volume of fluid de-livered to each distal nephron results not only ingreater CH2o per 100 ml GFR, but also in a higherrate of sodium excretion and a higher Uosm in wa-ter diuresis. The low absolute rates of CH20 mustbe due to a diminution in the total number ofnephrons.

The striking difference in the pattern of waterexcretion between acute and chronic hydronephro-

sis is displayed in Figure 7, where the fraction ofthe glomerular filtrate excreted into the urine(V/GFR) on the experimental side is comparedto that of the contralateral control kidney duringboth water and hypotonic saline diuresis. It isevident that at any given level of fractional volumeexcretion on the control side, fractional excretionis lower on the side with renal arterial constrictionor acute hydronephrosis and higher on the sidewith chronic hydronephrosis. The proportion ofthe GFR that is excreted into the urine duringmaximal suppression of antidiuretic hormone rep-resents the difference between fractional deliveryto, and fractional water loss from, the distalnephron. The low fractional volume excretionduring acute hydronephrosis and renal arterialconstriction is probably the consequence of dimin-ished delivery of filtrate to the diluting segment inthe face of a well-maintained absolute rate of freewater back-diffusion out of the collecting duct.The well-maintained absolute rate of free waterback-diffusion will result in an increased fractional

'loss of water and a diminished fractional volumeexcretion. The demonstration by Levinsky, David-son, and Berliner (20) that during antidiuresis

WATER DIURES/S

£a 0 0

so a 0°a

o0, l a

HI YPOTONICSALINE D/IRES/S

.

0 5 10 /5 20 0 5 /0 /5 20 25 30 35V

Cornzrol Xidnry GFiv X 100

FIG. 7. FRACTION OF GFR (V/GFR X 100) DELIVERED TO THE DILUTING

SEGMENTIN RENAL ARTERIAL CONSTRICTION AND ACUTEAND CHRONICHYDRO-

NEPHROSIS, COMPAREDTO THE CONTROLSIDE. Graph to the left represents dataobtained during water diuresis, graph to the right during hypotonic salinediuresis. Lines drawn are the theoretic lines of equal fractional delivery.

F

' 40o

X.0 35

b 25

'.. 20

'5

0

145

129

p

SUKI, EKNOYAN, RECTOR, AND SELDIN

renal arterial constriction sufficient to reduce GFRby 50% results in only a 15 to 20% reduction inmedullary sodium concentration supports the viewthat medullary hypertonicity is preserved despiteunderperfusion of the distal nephron.

The increased fractional volume excretion dur-ing chronic hydronephrosis could result from twofactors. Unquestionably the diminution in freewater back-diffusion secondary to medullary de-struction contributes to the augmented fractionalvolume excretion. A second contributing factor,also the consequence of medullary destruction,would be diminished loss of fluid from the descend-ing limb of Henle's loop (21) resulting in in-creased fractional delivery of filtrate to the dilutingsegment.

Clinically, chronic hydronephrosis is character-ized by a marked propensity towards polyuria withimpaired concentrating ability, which at timesreaches the proportions of frank nephrogenicdiabetes insipidus (1-5). This syndrome occurswith some frequency in hypocalcemic and hypo-kalemic nephropathy, but only rarely in parenchy-mal renal disease (1, 22-24). Since most formsof chronic renal insufficiency do not producenephrogenic diabetes insipidus, there must besomething distinctive in chronic hydronephrosisthat predisposes to this derangement. A plausibleexplanation may reside in the gross anatomicaldifferences between chronic hydronephrosis andother forms of renal disease. The renal paren-chyma in hydronephrosis of some duration isthinned out with flattening of the papilla and nar-rowing of the medulla, whereas in other forms ofrenal disease, medullary renal architecture is betterpreserved. The disproportionately severe destruc-tion of the medulla in chronic hydronephrosis re-sults in diminished medullary tonicity, which mayincrease the excretion of dilute urine by two de-rangements. First, the abstraction of water fromthe descending limb of Henle's loop will be greatlyreduced; the resulting augmentation of fractionaldelivery of filtrate to the diluting segment will in-crease free water formation. Second, free waterback-diffusion out of the collecting duct will begreatly reduced. The combination of increasedfree water formation and reduced free water losscould account for the increased frequency ofnephrogenic diabetes insipidus in chronic hydro-

nephrosis, as contrasted to other forms of pa-renchymal renal disease.

Summary

The effects of acute and chronic hydronephrosison renal function were investigated and comparedto those of renal arterial constriction during waterand hypotonic saline diuresis.

In acute hydronephrosis there was a reductionin urinary sodium concentration, sodium excretion,and free water clearance (CH2o) per 100 ml glo-merular filtration rate (GFR). The urinary os-molality was higher on the hydronephrotic sidethan on the control side during water diuresis andfell to below that of the control side during hypo-tonic saline diuresis. This pattern was similarto that of renal arterial constriction and, there-fore, was due to distal nephron underperfusion.

Chronic hydronephrosis resulted in a markedlydifferent pattern characterized by an increase inurinary solute concentration, urinary sodium con-centration, and sodium excretion; CH2O was de-creased, whereas CH2O per 100 ml GFRwas in-creased. This pattern is best explained by a com-bination of reduced renal mass with overperfusionof residual nephrons as a result of increased filtra-tion and diminished fractional reabsorption in theproximal nephron. Reduced back-diffusion offree water in the distal nephron may also con-tribute to the increased fractional excretion ofwater.

References1. Roussak, N. J., and S. Oleesky. Water-losing ne-

phritis: a syndrome simulating diabetes insipidus.Quart. J. Med. 1954, 23, 147.

2. Earley, L. E. Extreme polyuria in obstructiveuropathy. Report of a case of "water-losing ne-phritis" in an infant, with a discussion of polyuria.NewEngl. J. Med. 1956, 255, 600.

3. Winberg, J. Renal function in water-losing syn-drome due to lower urinary tract obstruction be-fore and after treatment. Acta paediat. (Uppsala)1959, 48, 149.

4. Dorhout Mees, E. J. Reversible water losing state,caused by incomplete ureteric obstruction. Actamed. scand. 1960, 168, 193.

5. Berlyne, G. M. Distal tubular function in chronichydronephrosis. Quart. J. Med. 1961, 30, 339.

6. Wilson, B., D. D. Reisman, and C. A. Moyer. Fluidbalance in the urologic patient: disturbances in therenal regulation of the excretion of water and so-

130

RENAL FUNCTION IN ACUTEAND CHRONICHYDRONEPHROSIS

dium salts following decompression of the urinarybladder. J. Urol. (Baltimore) 1951, 66, 805.

7. Bricker, N. S., E. L. Shwayri, J. B. Rearden, D.Kellog, J. P. Merrill, and J. H. Holmes. An ab-normality in renal function resulting from urinarytract obstruction. Amer. J. Med. 1957, 23, 554.

8. Van Giesen, G., M. Reese, F. Kiil, F. C. Rector, Jr.,and D. W. Seldin. The characteristics of renalhypoperfusion in dogs with acute and chronic re-ductions in glomerular filtration rate as disclosedby the pattern of water and solute excretion afterhypotonic saline infusions. J. clin. Invest. 1964, 43,416.

9. Kiil, F. The Function of the Ureter and Renal Pel-vis. Philadelphia, Saunders, 1957.

10. Malvin, R. L., H. Kutchai, and F. Ostermann. De-creased nephron population resulting from in-creased ureteral pressure. Amer. J. Physiol. 1964,207, 835.

11. Berliner, R. W., and D. G. Davidson. Production ofhypertonic urine in the absence of pituitary anti-diuretic hormone. J. clin. Invest. 1957, 36, 1416.

12. Ostle, B. Statistics in Research. Ames, Iowa, IowaState University Press, 1963.

13. Thompson, D. D., M. J. Barrett, and R. F. Pitts.Significance of glomerular perfusion in relation tovariability of filtration rate. Amer. J. Physiol.1951, 167, 546.

14. Dirks, J. H., W. J. Cirksena, and R. W. Berliner.The effect of saline infusion on sodium reabsorp-tion by the proximal tubule of the dog. J. clin. In-vest. 1965, 44, 1160.

15. Glabman, S., H. S. Aynedjian, and N. Bank. Micro-puncture study of the effect of acute reductions in

glomerular filtration rate on sodium and water re-absorption by the proximal tubules of the rat. J.clin. Invest. 1965, 44, 1410.

16. Brunner, F. P., F. C. Rector, Jr., and D. W. Seldin.Unpublished observations.

17. Kiil, F., and K. Aukland. Renal tubular localizationof water and sodium reabsorption in antidiuresisand water diuresis. Scand. J. clin. Lab. Invest.1960, 12, 277.

18. Gottschalk, C. W. Micropuncture studies of tubu-lar function in the mammalian kidney. Physiologist1961, 4, 35.

19. Bricker, N. S., P. A. F. Morrin, and S. W. Kime,Jr. The pathologic physiology of chronic Bright'sdisease. An exposition of the "intact nephron hy-pothesis." Amer. J. Med. 1960, 28, 77.

20. Levinsky, N. G., D. G. Davidson, and R. W. Berliner.Effects of reduced glomerular filtration on urineconcentration in the presence of antidiuretic hor-mone. J. clin. Invest. 1959, 38, 730.

21. Earley, L. E., and R. M. Friedler. Observations onthe mechanism of decreased tubular reabsorption ofsodium and water during saline loading. J. clin.Invest. 1964, 43, 1928.

22. Carone, F. A., and F. H. Epstein. Nephrogenic dia-betes insipidus caused by amyloid disease. Evidencein man of the role of the collecting ducts in con-centrating urine. Amer. J. Med. 1960, 29, 539.

23. Kleeman, C. R., and F. H. Epstein. An illustrativecase of chronic pyelonephritis with persistently hy-potonic urine. Amer. J. Med. 1957, 23, 488.

24. Kleeman, C. R., W. L. Hewitt, and L. B. Guze.Pyelonephritis. Medicine (Baltimore) 1960, 39, 3.

131