Renal glomerulotubular mechanisms during normal pregnancy

Vol. 75, No. 5 May, 1958

OBSTETRICS Tran.uctions

AND

GYNECOLOGY

of the Thirteenth Annuat

Meet&g

of the

Socie~ of 0bstetriGin.r and Gynae6ologi.s of Cn& RENAL GLOMERULOTUBULAR NORMAL PREGNANCY

MECHANISMS

DURING

I. Glomerular Filtration Rate, Renal Plasma Flow, and Creatinine Clearance* t RUSSELL

R. DE ALVAREZ,

M.D., SEATTLE,

WASH.

WITH THE TECHNICAL ASSISTANCE OF GLORIA E. BRATVOLD, B.S. (From the Medicine)

Department

of

Obstetrics

and

Gynecology,

University

of

Washington

School

oi

E

VER since it was first lea.rned that the kidney played a part in producing the clinical manifestations of pregnancy toxemia, this organ has continued to offer a major challenge to a full comprehension of the mechanisms involved in the genesis of this important complication of pregnancy. Since the kidney is the principal organ, exclusive of the uterus, to which attention is paid during abnormal pregnancy, it becomes all the more important, that a study of its physiology during normal pregnancy should receive special attention. Until the renal features present during normal pregnancy are better understood, it is not possible to develop any clear-cut approach toward a better understanding of the basic renal alterations which occur in the toxrmias of pregnancy. *Supported, in part, by a grant-in-aid from the United States Public Health Service and from Initiative 171 Funds. pAddress of the Guest Speaker at the Thirteenth Annual Meeting of the Society of Obstetricians and Gynaecologists of Canada, Banff, Alberta, June 22, 23, and 24, 1957. !I31

Since renal function cmbc)tlies the l~roc~~sscs cnrric~l out by I he Iticlrleg ill t,hc excretion of waste products, sob&, mct,abc~Iitrs, and electrolytes ill \v:l.t(xr, further knowledge of these processes becomes necessary. T~~as~nucl~ :u+ the transport of these products to the kidney is carrictl out by t,hc systemic circulation, the maintenance of normal renal function obviously depends upon vascular integrity. The maintenance of the hcmodynamics involved in normal renal function most probably depends upon a monitor system, a system of intcgration, and one to maintain normal homeostasis. It has long been understood that the formation of urine results from glomerular filtration; Ludwig1 was probably the first to employ this thesis. Huber’ further advanced the knowledge of renal function by being the first to dissect, out and to reconstruct the complete nephron. Combining these two advances, Cushny3 believed that the formation of urine resulted not only from the filtration at the glomerulus but that tubular reabsorption of this glomcrular filtrate also occurred. It was not until Richards” proved this hypothesis that the present-day knowledge of glomerulotubular mechanisms was attained. He was the first to demonstrate that all the products contained in the plasma water also were present in the fluid obtained from Bowman’s capsule after allowing for the Donnan equilibrium. Although numerous renal function tests have been devised, the majority are not capable of measuring discrete and specific renal functions in quantitative terms. Without quantitation, the part played by the kidney in water and The commonly employed tests, electrolyte metabolism tends to be clouded. such as urea clearance, phenolsulfonphthalein clearance, dilution, and concentration tests, and the determination of blood and urinary metabolites, do not offer a truly quantitative evaluation of renal function. The generally available laboratory and clinical techniques used for determining renal function in clinical pract,ice can at best bc classified as only rough approximations of renal function. The introduction of improved techniques, particularly those proposed by Homer Smith,” now makes it possible to perform accurate determinations of renal function, as well as to explore the renal physiology as it relates to the obstetric patient. Clomerular filtration rate may be measured by inulin clearance while the renal plasma flow (as well as renal blood flow) may be calculated in its relationship to effective tubular tissue. Renal plasma flow is calculated by determining the clearance of para-aminohippurate (PAH) at low plasma levels. This permits rational evaluation of kidney function and also allows calculation of the measure of excretion or reabsorption by the tubules. From these, the fraction of the glomerular plasma filtered through the glomerular basement, membrane can also be calculated. Not only may plasma fluid filtered at the glomerulus be measured, but the electrolyte filtration at the glomerulus can also be calculated, thus permitting evaluation of the renal mechanisms of clcctrolytc metabolism. While most studies of renal function in the pregnant patient have been performed on patients with pre-eclampsia-eclampsia, specific renal disease, or

Vdilme Number

75 5

RENAL

GLOMERULOTUBULAR

MECITSNIPMS

TN PREGNANCY

!)33

hypertensive cardiovascular disease, only sporadic reports have dealt with renal mechanisms in the normal pregnant patient. The majority of these Iatt(lI* studies have been “spot ” studies performed at or near the end of prcgnaney. Less than a handful of investigators have studied renal physiology in respct*l to the specific renal functions of glomcrular filtration ant1 renal plasma tlow. In somt investigations where correlations have been made in thtl same patient. only the changes of late pregnancy have becu correlated with those which occurred during the early pucrperium. Some authors, like Welsh and associat CS,” conclude from studies of this type that no significant altera.tions in the renal hemodynamics occur in normal pregnancy. Bonsnes7 st,ated that th(l glomcrular filtration rate was increased by approsimatclp 50 per cent nbovc~ the nonpregnancy level of 120 cc. per minute to a level of 180 c.c. per minut(~ a’t the tenth week of pregnancy. Although his report. did not tabulate or illustrate his tlata, he stat,ed that these levels persisted throughout pregnancy nnfl returned to the nonpregnancy level of 129 C.C.per minute at the thirt,y-nint,h week of pregnancp. He recorded no postpartum studies. In 1951 Bucht.” ut,ilizing a single intramuscular injection of PAH and a single intravenous administration of 90 C.C. of 10 per cent inulin, studied kidney function in nonpregnant, and normal pregnant women, divicling the latter group into two separ;ltc subgroups for comparison. The first subgroup was composccl of patients with a duration of pregnancy extending through the eighth lunar month ; the second subgroup consisted of patients who were in t,he last two lunar months of pregnancy. These two subgroups were compared with each other. He also reported an incrpasc in glomerular filtration rate. Soharg showed that thr glomerular filtration ra.te followed the same pattern as that described by Bonsnes. Sims and Krantz,l” studying 12 pregnant subject,s, mostly by serial evaluations beginning at the sixtrenth week of pregnancy, found the glomerular filtration rate to hc increased t.o 166 C.C. per minute and to persist at this level until the thirt,yeighth week, at which time it dropped to 146 C.C.per minute for the remainder of pregnancy. The glomerular filtration rate in t,hcir patients dropped to !G c.e. per minute late in the puerperium. These latter invcst,igators also notctl that the renal pla.sma flow was significantly elevated from the sixteenth wcrk of pregna.ncy, with levels of 820 C.C.per minute. dropping to 590 C.C.per min.utc just before term. The renal plasma flow further dropped t,o below the normal nonpregnancy range for several weeks in the postpartum period hcforcl eventually returning to the normal. While their report did not show the esact nature of the curve, Chcsley,ll in a personal c~nni~li~~Iii(.ati~)lito us, I+ ported the trend quoted above. Since the features of renal function arc dependent upon t,hc general (:itaculatory hemodynamics, then the renal hemodynamics also depend upon alterations in the circulation during normal pregnancy. The renal fraction cornprises approximately 25 per cent of the total cardia.e output. It is a wcllknown fa,ct that the circulating plasma volume increases by approximately : confirm but augment the multiple previous stud& done with the use of thtr

dye techniques. While the blood pressure dots not change significantly during pregnancy, a slight reduction in diastolic pressure does occur. C’ar(liac output reaches its maximum between the twenty-sixth and thirtieth weeks of pregnancy and gradually returns to normal after the thirty-second week, it appears that, since the cardiac output does increase significantly during prcgnancy, there might be other requirements for this increased cardiac output over and above those required by the fetus. It could well be that the requirtmcnts of the maternal kidney to excrete waste products rcachcs its maximum at this time, or that an increase in hemodynamic force is required to propel the plasma wat.er through the capillaries of the nephron.

Subjects and Materials In an attempt to develop information as to just what happens to renal function in normal pregnancy, serial determinations of glomerular filtration rate, renal plasma flow, filtration fraction, and creatinine clearance were performed on a group of normally pregnant subjects. Twenty-two patients comprised the total series. Fourteen patients were initially selected for the serial study. All patients were ambulatory and were carrying out their usual normal activities. None .showed evidence of heart disease, renal disease, or hypertensive cardiovascular disease, nor was any patient included in the series in whom clinical manifestations of toxemia developed. Nine normal pregnant patients completed the study and were followed at varying intervals from the seventh week of pregnancy on. In most instances the investigations were performed every 4 weeks through the prenatal period, on the third postpartum day, and again at 6 weeks post partum. Four patients who did not complete the total study had two or more renal function studies performed during the pregnancy ; 2 of these pa.tients had abortions, the husbands of the other 2 were transferrecl from this area. One patient is continuing serially at the present time. Isolated single renal studies were performed on 8 additional patients. None was receiving medication other than the usual prenatal supplements. All patients were on the usual prenatal diet for normal pregnancy without stringent restrictions, with the possible exception of calories. Since none of these patients was a hospita,l patient, it was felt that the data obtained on these patients might bc more consistent than data obtained from patients kept at prolonged bed rest, hospitalized for complicating features, or from patients who had had disease processes which had since returned to normal.

Methods A total of 85 clearance studies was performed. All patients were fasting at the time of each determination. Each patient received 600 cc. of water orally 2 hours prior to the beginning of each test. A 24 hour urine specimen was collected during the 24 hours immediately prior to the determination of renal hemodynamics ; this 24 hour specimen was procured to measure creatinine clearance. At each visit, simultaneous inulin, PAH, and creatinine clearance values were obtained. The filtration fraction was also calculated. At the times of the actual determination the bladder was catheterized and the bladder drained through a multiple-opening catheter which was left inlying for the collection of subsequent urine specimens. The first urine specimen collected was discarded. The first heparinized blood sample of 12 CC. was taken through a No. 18 needle and was used for blanks in subsequent chemical determinations. This needle was kept in the vein and the tubing

Volume

75

Number 5

RENAL

GLOMERULOTUBULAR

MECHANISMS

IN

PREGNANCY

!R-j

from the infusion pump attached, through which a priming dose of 30 C.C. of 10 per cent inulin was then introduced intravenously, to bring the body to a state of equilibration. The sustaining solution of inulin, para-aminohippur~,t~~ and 5 per cent dextrose in water was then administered through the samc~ needle, attached to a 3 way stopcock. The illfusion was sustained at a constant rate of 1 C.C.per minute, to maintain a controllctl J~~SHI~ level throughout the testing periods. Twenty minutes la,ter, a No. 18 needle was placed in a vein in tht opposite arm, with a 3 way stopcock inserted between the glass adapter and the needle for the removal of blood specimens. The first blood sample was takrn at this time. At the same time the bladder was drained; each collected urine spccimen was measured and an aliquot saved for laboratory analysis, thus initiating the first clearance period. This same procedure was rep&cd every 20 minutc%s for a total of four t,o six periods ( a span of al)J)rosimatcly 2 hours). Blood pressure and pulse rate were determined and recorded. The exact time of collection of each specimen was rccordcd anti a complctc notation of the patient’s reaction and any unusual cvcnt,s in the procedure were notc*tl. The number of cubic centimeters of sustaining snlution injected was notctd for each period (a period consisted of the time from the cinclof one urine collcc+ljrt through the end of the next urine collection). The rate of glomerular filtrat.ion was measured by innlin clcara.nce. Tnulin concentration was measured in protein-free filtrate of plasma and in diluted urine, by the resorcinol method of Schreincr.l.* Inulin, a polvsaccharide, is neither reabsorbed nor excreted in t.he tubule. When introduced”intravenonsly, it is soluble in the plasma water, does not combine with the plasma prot,eins, is filtered with the plasma water at the glomernlus, and is neither added to nor removed from the renal tubular fluid. Its clearance is equal to the rata of filtration; thus all inulin filtered at t,he glomcrulua is recovered in the urine.. Hence, the amount of inulin found in the urine equals the amount of inulin filtered and is the same amount in the plasma watr~~ as is filterecl through thtl glomeruli in one minute. The rate of glomerular filtration can thus bc tlctermined, since the amount and concentration of innlin RW known in one minntc’s filtrate. The renal plasma flow was measured wit,h para-aminohippurate (PAH) by the method of Smith.15 A certain portion of the tot,al amount. of PAR injectctl is bound to the plasma protein while the remainder is in free?solution in pl;lxn~a water with which it can pass through the glomcruli with t,he glomcrular filtrate. At the proximal convoluted tubule, PAH cntrrs the ceJJ and is transferred through it by a physicochemica.1 mechanism. Tt is then secreted by the tubules into the tubular fluid and excreted in the amount equal to the conccntration in the plasma water as the circulat~ing pla.sma, passes through th(a tubules during the period of excretion. PAW, at low plasma levels, is pramtically completely removed from the renal arterial blood in one circulation through the kidneys and the clearance of thcsc substances is identical with the rate at which blood itself is circulated through the kidney. Thus, t,hcn renal excretion of PAR occurs not only by glomcrular fiJtra.tion but also by tubular excretion. Endogenous creatinine was measured in scru m iltltt nrinp by moclifica t.icms of the method described by Pe&rs.16

Results Glolnerular B’dtration Rate.--Fig. 1 shows the values of the glomerular~ filtration rate found in all patients studied throughout pregnancy. Not only are the actual values of each test depicted, but the mca,u value and trend as

936 each individual pregnancy advanced are also graphically represent,ed. The trend of a progressive decrease in the gIomeruIar fiItration rate as pregnancy continues becomes more obvious when these average values are studied. In order to determine possible differences in values among patients not consecutively evaluated, the serial study group is augmented by additional pstient,s comprising a group on whom random “spot” studies of renal hemodynamics were performed. Fig. 2 shows the average values obtained in the pat,ient,s studied serially from early pregnancy through the late puerperiurn and a comparison of our values with those reported by other authors. These values arc also compared with the normal nonpregnancy level of 108 C.C. per minute. As early as the seventh week! the glomerular filt,ration rate was elcvatetl to levels as high as 167 cc. per mmute. This represents an increase of 60 per cent above the nonpregnancy rate. During the latter part of the first third of pregnancy, the glomerular filtration rat.c begins a steady decline, yet, rcmains above the normal nonpregnancy level during this trimrster. GLOMERULAR

FILTRATION RATE

[NORMAL

PREGNANCY)

.

.

:

D----O 0'

4

Serial A “Spot” Average of Semi Average of Serial

8

Fig. nancy.

.

::

.

l

)

/ / plus “Spot”

I2

16

20 WEEKS

represents

l

:

. l

24

28

32

36

OF GESTATION

l.-Glomerular filtration rates determined during normal pregnancy. the group of patients upon whom serial determinations were done The dotted curve represents the average of the serial and the “spot”

40 Lessftl.3” 6 Wtxkl

6 0.w PO61 PO,l”rn I POLf Furturn

The solid line through pregpatients.

The decrease in the rate of filtration of plasma water at the glomerulus continues during the middle third of normal pregnancy, eventually attaining values somewhat below those of the normal nonpregnant female, and persisting thus until the time of delivery. After delivery the glomerular filtration rate rises abruptly, reaching an average value of 130 cc. per minute by the third postpartum day. This increase is associated with and is possibly responsible for the postpartum diuresis. By the sixth week post partum, the glomerular filtration rate has returned to levels approximating those found during the last trimester of pregnancy. Renal Plasma Plow.-Renal plasma flow as measured by para-aminohippurate clearance in the 9 serially studied patients and the “spot” determinations are illustrated in Fig. 3. The average values of the entire group are presented in Fig. 4 and are compared with the values of other authors. Here, too, high values of renal plasma flow were attained in early pregnancy, also 60 per cent

This lat,tc:r valt~l~ above our normal nonprcgnancy level of 599 c.e. per minute. coincides almost esaetly with that obtainetl by ltomer Smith.” These total> elevations persist during the first trimester Of pwgnancy, then gradually tit*‘l’lll(. dine to the normal nonpregnancy range during the middle trimester. progressive depression in renal plasma flow continues! even though within I hrt normal nonpregnancy range, until about t,he thirty-sixth wwk of pr(~gn~~n~~y. at, which time the values fall to slightly bc11bw t,hc avrrage figures for the Ilot’anal nonpregnant pat,ient. In a pattern similar to that of the glomerular filtration rate, renal plasnlil flow rises during the early postpartum period. reaching normal levels by tht! filtration rate, the renal plasma t,hird postpartum day. Unlike the glomcrular flow not only attains normal nonprcgnancy levcxls by the t,hird postpart,um (1;~ but also remains at essentially the ~~orn~al nonpregnancy 1~~~1thrtrnghont I II+% rritirr pucrperium. GLOMERULAR

FILTRATION

(COMPARATIVE

RATE

VALUES)

2001

I I

I

,““““~~..........‘......’

..“....“““.........*.,

___-

Normal Pregnant )c-- ---x Welsh, Welien,Toylor m--.-m Eonnner and Lange +.--.* Bucht, Ii. ----a Sohor, et 01. C - --A Sims and Krontz e----O deAlvorez

Normal -----

Non-Pregnant Smith, H.W. deAlvarez

I 0’

8

16 WEEKS

by tiOn

24 OF GESTATION

32

40 L*Ds m m 6 web 6 cloya PO*+ PWfum i Post Fwtvm

Fig. 2.-A comparison of our values for the glomerular filtration rate with those other authors. It is noted that our figures show a progressive decline in glomerular rates and that all flgures for the most part, are lower than those usually quoted,

stated flltrs-

B’iltration Fraction.-The values of the filtration fraction, representing the ratio GFR/RPF, are demonstrated in each individual pregnancy in Fig. 5. The average values for the serial, as well as for the total group illustrate a remarkably consistent trend as pregnancy advances. During early pregnancy, the filtration fraction assumed values lower than normal, reaching levels as low as 14 per cent; the reduced values persisted until the latter part of the second trimester. Normal nonpregnancy values were attained early during the last trimester of pregnancy and maintained throughout the remainder of pregnancy. At the third postpartum day, the filtration fraction had continued to rise, reflecting the increase in glomerular filtration. Creatinine Clearance.-During the 24 hours immediately preceding each test of inulin and PAJ3 clearance, all urine was collected for creatinine clearante; a simultaneous blood specimen was also taken. Even though creatininc

I’ABLE .__

-~_.

-_-~--.-_. ~-___-.

~~~

GROI: 1’ I TRIMESTER (O-16

FIRST NO.

(c.e./min.

Creatinine clearance min./1.73m.z) Renal plasma minJ1.73m.2) Filtration

flow

(cc./

OF

8TATlSTlCA1,

11

14

10

13

WEEKS)

i2iYAI,YSIS

GROUP TRIMESTER

SECOND

I

I 1 j

/ DETERMI1 NATIONS

PATIENTS

inulin clearance /1.73m.z)

1.

NO.

II (l’i-28

OF

I

fraction

(

rO)

11

14

I1

14

DETERMINATIONS

S

133.2 t4.28 112.s r)

14.2

17

34

108.2 24.57

23.5

17

34

107.7 26.62

17

34

731 539.4

17

34

986 238.0

PATIENTS

(NORMAL

126 1.2 1 I

13.80 10.36

RENAL PLASMA 1400

MEAN

15.80 lrl.04

FLOW

PREGNANCY)

I200

. . .

1000

*

:

.

400 l

I

o-----l3

200

0

Serial

4

Fig. 3.-Renal line represents the combination of the done. A progressive of pregnancy.

. .

.

. ‘“spot” Average Average

of Serial of Sellal

8 plasma average patients decline

2

.

.

’ I , /

plus “spot”

I2

I6

flow is of the studied beginning

k.XAI,

20 WEEKS depicted patients serially early

i4 ie OF GESTATION for all serially as well

in

pregnancy

patients studied. as those

3.2

followed The upon

continues

40” LCSL man 6 vie*5 ; 6 Doyl PO9 Porfun , POLl mrrum

j6

in broken whom throughout

Our

study. curve “spot” the

The represents studies total

WEEKS‘)

L-

I

MEAN

27.44 (c.c./

OF

__-

solid a were length

clearance does not provide as accurate or as uniformly quantitative information relative to glomerular function as does inulin clearance, it does give some information as to the trend of renal function during pregnancy. The values as well as the trend are noted in Fig. 6. It is noted that increased creatinine clearance occurs during early pregnancy, followed by a subsequent reduction. During the remainder of pregnancy, the pattern of creatinine clearance was similar to that of inulin clearance. In the puerperium, both early and late, the trend of the two curves was identical.

I s ---

18.9

37.3 163 4.3

~;‘o;;~;S HEMODYNAMICS

RENAL

DURING

GLOMERULOTUBULAR

NORMAL

MECH-%NISMS

IN

THIRD

NO.

III ~29-40

LESS

WEEKS)

NO. S

104.9 T8.93

26.8

20

101.4 25.93

17.8

20

558 250.5

20

-_

19.52 k1.12

3.4

IV POST

PARTUM

OF

-______-..-.-. I

MEAN

i

S

i

GROUP

6 WEEKS NO.

.-.-

DETERMINATIONS

I ‘ATIENTS

151

GROUP 6 DAYS

THAN

OF DETER&CINATIONS

PATIENTS

!XS!)

PREGNANCY ~-_^--__-.

GROUP TRIMESTER

PREGNANCY

PATIENTS

--

v-------

POST

PARTDhI ,

OF DETERMINATIONS

’ MEAN

! /

--_

S

10

10

129.5 k2.27

7.2

7

7

96.7 k5.59

14.X

9

9

107.1 25.68

17.0

7

7

73.7 27.3 1

19.6

10

10

7

7

10

10

i

7

598 t32.9

704

22.23 21.30

4.1

605 +45.0

119

15.83 kO.59

I.6

RENAL PLASMA FLOW (COMPARATIVE

VALUES)

1200

1000

SO0 I? w . $ : 2

600

400

200

X------Q h---.4 c -- -4 O-----o

Normal Prtqnm Welsh, Wtlltn, Taylor Bucht, H. Sohar, al. Sims ond Krantt deAlvow

Norm01

-----

et

0 El

Non-Pwnant Smith, H.W. dtAlvartz

16

24 WEEKS

Fig. 4.-Renal plasma flow as determined the literature. While most authors indicate curve for the various durations of pregnancy

32

OF GESTATION

by that has

Y

I I 1 I t I I

40 L~SI than

6 Wcek~ I 6Dey1 Post Por~ur 1 PaIf PD,lYrn

us is compared with the values a drop in renal plasma flow not heretofore been presented.

reported occurs,

in the

Statist&al Analysis.-All data were subjected to statistical analysis and are presented in Table I. The number of patients and number of determinations are noted for each type of clearance, including mean, standard error of the mean, and standard deviation. When further analyzed by the 3” test, a comparison of the first-trimester values of the glomerular filtration rate and the renal plasma Aow with the values obtained in the second trimester, third trimester, and 6 weeks post partum reveals the differences to be statistically significant. Similar comparisons of creatinine clearance show no statistically significant differences, however.

DE

940

Am. J. Ok.

ALVAREZ

FILTRATION (NORMAL

FRACTION

PREGNANCY)

I

35l

30-

& Gynec.

May.1958

I

Serlol

I I /

6 ‘spot’ Average of Serial Average of Serial

-

.

plus “Spot”

.

25> 'u 202 211 15-

/

.

t

l

1:

.

.

I I I

I I , I

502

4

8

12

16

20 WEEKS

Fig. 5 .-The renal plasma flow) curve indicates a “spot” studies were A progressive

24

26

32

36

h"

&Ler.

/ ,,~,O,“,‘:,,,

, 6 w&ir

Post Parturn

OF GESTATION

flltration fraction (the relation of the glomerular filtration is averaged in the solid curve for the patients studied serially. combination of the patients followed seriahy and the group performed. rise occurs as term is approached.

rate to the The broken upon whom

CREATININE CLEARANCE (NORMAL

PREGNANCY)

200

..L-u/ .

I50 F: -2 * $ 100 .$ k ib

. l

. . .

\

”

. .

50 W D----a

0

l. ?G . :Q

l

Serial

.

“Spot” Average Average

4

8

of Serial of Serial

12

.

” .

plus ‘Spot’

16

20 WEEKS

Fig. 6.4reotinine represents the group nancy. The dotted

. .

.

24

28

32

36

1

OF GESTATION

clearance rates determined durin normal pregnancy. The solid line of patients upon whom serial determ Lb tione were done throuerbout pregcurve represents the average of the serial and the “spot” determinatfons.

Even though the observations made by us do not permit detiite statements as to the cause of the mechanisms involved during normal pregnancy, the results do indicate the pattern at varying intervals during the same pregnancy in the same patient. The precise alterations in each mechanism depend

$d;hmr’,”

RENAL

GLOMERULOTUBULAR

MECHANISMS

IN

PREGNANCY

!I‘41

upon so many simultaneously operating variables that their explanation needs continued investigation. An increase in systemic blood pressure constitutes a plausible explanation for an increase in glomerular filtration rate in some cardiac abnormalities ; however, no such increase was noted in our patients, nor is a rise in mean arterial blood pressure a feature during normal pregnancy. While an increase in venous pressure does occur in normal pregnancy, it does not do so until late pregnancy. Antecubital and femoral vein pressures have been shown to be normal during early pregnancy. Even though it has been shown that there is no increase in peripheral venous pressure, it is not known whether venous capillary pressure is increased during normal prcpnancy. Possibly the increase in the glomerular filtration rate during early pregnancy might most logically be explained on the basis of dilatation of the afferent glomerular arteriole, with resultant increase in intraglomerular prcssure. Mild constriction of the efferent glomerular arteriole might also btx postulated as the basis for an increase in glomerular filtration. On the basis of our investigation this appears to be a temporary phenomenon, inasmuch as the glomerular filtration rate does not remain elevated for any prolonged period of time. The progressive decline in the glomerular filtration rate during normal pregnancy, as noted among our subjects, does not lend it,self to easy explanation. Since the elevation in the glomerular filtration rate occurs at a time when the level of circulating chorionic gonadotropin is highest, it might be postulated that the increased glomerular filtration rate in early normal prcgnancy is on this basis. As the titer of the chorionic gonadotropin decreases. a simultaneous increase in circulating steroids occurs during normal Dreg‘nancy. It is now known that, among these, the 17-hydroxycorticosteroids also are elevated as pregnancy advances. It is generally considered that these a.re of placental origin and may also play a role in renal vascular changes during pregnancy. With the decreased systemic blood pressure which occurs in most pregnancies a reduction in filtration pressure probably occurs simultaneously. Even though this change in systemic blood pressure is not significant, it is know11 that the pressure in the peripheral vascular bed is reduced during normal pregnancy. Also during normal pregnancy an increase in peripheral circulation occurs so that possible compensatory afferent arteriolar vasoconstriction may occur. If renal vasoconstriction occurs, it is likely that it would be generalized and that efferent arteriolar vasoconstriction would occur simultaneously. To cause such degrees of vasoconstriction, however, other cvidencp of arteriolar vasoconstriction should be present but this is not found in nor-. ma1 pregnancy. Furthermore, as the requirements of the placenta and thtp peripheral circulation create increased demands on the cardiac output. it is conceivable that shunting of blood from the kidney occurs and that eff(brent glomerular vasoconstriction attempts to maintain homeostasis. The high renal plasma flow noted in early pregnancy might best be es. plained on the basis of an increased cardiac output during pregnancy, ‘Ihe

942

IDE

ALVAREZ

Am. .T. Obst.

% Gym. May. 1958

curve of cardiac output has been plotted for normal pregnancy with the USC of dye injection methods; even bhe more recent direct methods of determining cardiac output have shown the curve to be similar. Later in pregnancy, particularly during the middle trimester and especially about the twenty-eighth week, it seems that the needs of the peripheral vascular bed and the fully dcveloped placenta account for the known increased cardiac output. These requirements probably account for the reduction in the renal fraction at this time. Inasmuch as the needs of t,he placenta, fetus, and the peripheral circulation continue ant1 the cardiac output begins to decrease during the last trimester of pregnancy, these could well account for thr progressive dccrcasc in renal plasma flow as noted among our group. During the first two trimesters of pregnancy the filtration fraction remains low. The increase in filtration fraction which occurs in the last t.rimester of pregnancy must imply increased tubular activity. While the glomerular filtraCon is maintained at a fairly constant level during the last third of pregnancy, renal plasma flow continues to decline. Analysis of these fact,ors in normal pregnancy seems to indicate that the increased fraction of the circulating plasma filtered at the glomerulus results in a proportionate increase in tubular reabsorption of water and electrolyte.

Summary When compared with existing data, our findings of renal glomerulotubular patterns during normal pregnancy seem to be in conflict with those cited in most reports. Perhaps it is significant to note that most of the existing literature does not deal with serial determinations of renal hemodynamics performed in the same paCent at different intervals throughout the same pregnancy. The csplanation that apprehension during the first examination is responsible for the elevated glomerular filtration rate cannot be confirmed, since the values obtained among the “spot” determinations were comparable to those obtained from patients studied serially. All our patients were ambulatory, were carrying out their usual normal activities, and were not placed on restrictive diet, or intake. The constant-infusion technique was used throughout and a uniform concentration of the testing agents maintained by intravenous administration of all testing agents. Diuretics or other artificial mcans to produce conditions different from those ordinarily seen in the normal pregnant patient. were not utilized. The reduced glomerular filtration rate and renal plasma flow in the normal pregnant patient may well permit the increased tubular reabsorption, not This then appears to establish a perfeet only of water but of sodium as well. prelude for an increase in the sodium space and therefore for the development of toxemia of pregnancy with its characteristic edema. It seems that, from these studies, most patients could well be on the brink of developing some degree of pre-eclampsia during what is supposed to be an apparently normal pregnancy.

Volume Number

75 5

RENAL

GLOMERULOTUBULAR

MECHANISMS

IN

PREGNANCk‘

!M

Conclusions 1. During the first trimester of pregnancy, the glomerular filtration rate is increased by 50 to 60 per cent above the normal nonpregnancy level. 2. Beginning early in the second trimester of pregnancy, the glomerular filtration rate progressively declines, gradually reaching a level below that of the normal nonpregnant patient. 3. Throughout the last trimester of pregnancy, the glomerular filt,ratiorl rate is maintained at a level slightly below the average level of the normal nonpregnant patient. The decrease in glomerular filtration rate during the sectonrl and third trimesters is statistically significant when compared with that 11-11 the first trimester. 4. A sharp rise in glomerular filtration rate occurs at the third postpartuln da.y, but by the sixth week post partum the glomerular filtration rate has VPturned to the level found at term. 5. During the first trimester of pregna,ncy renal plasma flow is olCv:itCIl 60 per cent above the normal nonpregnancy level. 6. During

the second trimester

of pregnancy,

a progressive

reduction

ill

renal plasma flow occurs, even though the values remain above those! of t.hd. average nonpregnant patient. This fall begins at the onset of the scconti third of pregnancy and continues to decrea.se until term. From the t,hirtieth week to term, the values for renal plasma flow remain consistently below the Renal plasma flow rc%lrtrs average values of the normal nonpregnant patient. to the normal nonpregnancy level by the thircl day post parturn anal remains so t,hrough 6 weeks post partum. 7. The filtration fraction is reduced during th(l first two trimest,crs of prcg~ nancy but increases during the last trimester. 8. Creatinine clearance increases only during the first trimester of pregnancy. This elevation is not sustained; the values drop and remain at, level!: slightly below the normal nonpregnancy level throughout the remainder of pregnancy. Even though the trend of creatinine clcaranee during pregnancy follows much the same pattern as that of inulin clearance, a comparison oi the results obtained during one trimester with those of other trimesters of pregnancy reveals no statistically significant diffcrcncc>. Creatininc clearancct. on the basis of our findings and when compared with the more accurate inulin clearance, cannot be considered as a reliable measure of the changes in ~~t~n;ll hemodynamics during normal pregnancy. 9. Renal function, as studied by us, does undergo changes during normal pregnancy

and in the earIy puerperium.

These alterations

in renal function

seem to be related to the sodium and water rcltention known to occur in normal pregnancy. Because the changes in renal function during normal pregnancy can bc related to accompanying hormone changes, it is suggested that. the changes in renal hemodynamics ilnrin g normal pregnancy arc tncdintcc-I by humornl influences.

944

DE ALVAREZ

Without the interest and untiring assistance of Lieutenant-Colonel Walter M. Wolfe, Chief of Obstetrics and Gynecology, Fort Lawton Hospital, Seattle, Washington, in securing and scheduling patients, this study would not have been possible. We wish to express our sincere appreciation to Miss Donna Simkins and Xiss Janice Ekholm for additional technical assistance; to Mr. Shashanka S. Mitra and Dr. Blair M. Bennett for assistance in statistical analysis; and to Miss Jessie Phillips, Head of Medical Illustration, and her staff, for medical illustration and photography. We are deeply grateful to Mr. Robert E. Parcell, Vice-President of Arnar-Stone Laboratories, Inc., for supplying inulin used in the study, and to Dr. James M. Sherwood and the Division of Clinical Research of Merck Sharp & Dohme for the supplies of paraaminohippurate. Their kindness and generosity made possible the execution of this study.

References 1. Ludwig, C.: Wagner’s Handb.d.Physiol. 2: 628, 1844. 2. Huber, G. C.: The Morphology and Structure of the Mamma1ia.n Renal Tubule, Harvey Lecture 100. 1909-1910. 3. Cushny, A. R.: ‘The Secretion of the Urine, 1917, Longmans, Green, & Co. 4. Richards, A. N.: Processes of Urine Formation, Proc. Roy. Sot., London, s.B 126: 398, 1938. 5. Smith, H. W., Goldring, W., and Chasis, H.: J. Clin. Invest. 17: 263, 1938. 6. Welsh, Catherine A., Wellen, Irwin, and Taylor, H. C., Jr.: J. Cbn. Invest. 21: 57, 1942. 7. Bonsnes, Roy W., and Lange, W. A.: Federation Proc. 9: 154,195O. 8. Bucht, H.: Seandinav. J. Clin. & Lab. Invest. (supp. 3) 3: 1,1951. 9. Sohar, Ezra, Scadron, Eugene, and Levitt, M. F.: Clin. Res. Proc. 4: 142, 1956. 10. Sims, Ethan A., and Krantz, Kermit E.: Clin. Res. Proc. 4: 142,1956. 11. Chesley, Leon C.: Renal Excretion of Water, Urea, Sodium, Potassium, Chloride and Protein, Ross Laboratories, First Obstetric Research Conference, Toxemia of Pregnancy, April, 1956. 12. Rose. David 5.. Bader. Mortimer, Bader, Richard, and Braunwald, Eugene: AM. J. 'OBET.& Gi~~c.721 233.1956.' 13. Hendricks, Charles H., and ‘Quilligan, Edward J.: AM. J. OBST. & GYNEC. 71: 953, 1956. 14. Schreiner, George: Proc. Soo. Exper. Biol. & Med. 74: 117, 1950. 15. Smith. Homer W.: J. Clin. Invest. 24: 388. 1945. 16. Peters, ’ -~~ J. H..ma J. Biol. Chem. 146: 179, 1942: of Renal Physiology, New York, 1956, Oxford University 17. Smith, H. W.: Principles Press.