RELATIONS OF EFFECTIVE RENAL BLOODFLOW AND GLOMERULAR FILTRATION TOTUBULAR EXCRETORY MASS IN NORMAL MAN

William Goldring, … , Hilmert A. Ranges, Homer W. Smith

J Clin Invest. 1940;19(5):739-750. https://doi.org/10.1172/JCI101179.

Research Article

Find the latest version:

http://jci.me/101179-pdf

RELATIONS OF EFFECTIVE RENAL BLOODFLOWAND GLOMERULARFILTRATION TO TUBULAREXCRETORYMASS

IN NORMALMAN1

By WILLIAM GOLDRING, HERBERTCHASIS, HILMERT A. RANGES, ANDHOMERW. SMITH

(From the Departments of Physiology and Medicine, New York University College of Medicine,and the Third (New York University) Medical Division of

Bellevue Hospital, New York City)

(Received for publication May 13, 1940)

In 1938 Smith, Goldring and Chasis (12) de-scribed methods for measuring the "effectiverenal blood flow" (diodrast clearance) and the" tubular excretory mass " (diodrast Tm) by theexcretion of diodrast at low and high plasma lev-els. The present report concerns the applicationof these methods to normal subjects under stand-ard or quasi-basal conditions, with special ref er-ence to the quantitative relations between the dio-drast clearance, the inulin clearance and diodrastTmin various subjects.

The subjects examined here were convalescent,volunteer patients of the Third (New York Uni-versity) Medical Division of Bellevue Hospital,and were selected as carefully as possible to ex-clude renal disease. Clearance determinationswere carried out in the morning without break-fast, although in a few instances a glass of orangejuice was allowed. The subject was hydrated bythe administration of one liter of water the nightbefore; the water was measured out for him andhe was allowed to take it ad libitum and withoutsupervision. The next morning a second liter ofwater was taken between 7:00 and 8:30 a.m., andthe subject was allowed out of bed momentarily.At about 9:00 a.m. an intravenous infusion con-taining the requisite concentrations of diodrast,

1 This investigation has been aided in part by a grantfrom the Commonwealth Fund.

The authors are indebted to the Winthrop ChemicalCompany for generously supplying some of the diodrastused in the later observations, and to Messrs. Hynson,Westcott and Dunning for the preparation of phenol redin sterile 10 per cent solution. They also wish to ex-press their appreciation of the continued interest of thePfanstiehl Chemical Company in the preparation of pureinulin. Recently, the U. S. Standard Products Company,Woodworth, Wisconsin, with the co-operation of Dr.Alf S. Alving, have made physiologically tested inulinavailable in sterile ampules.

phenol red and inulin in 0.9 per cent saline wasstarted, and urine collections were begun after asuitable washout period. Urine was collected bycatheter, the bladder being rinsed with saline atthe termination of each urine collection period.Sterile technique was observed throughout.

DIODRAST CLEARANCE

It has been our custom after 3 or 4 periods,during which the diodrast clearance has beenmeasured at low plasma levels of diodrast (0.8 to2.0 mgm. per cent of diodrast iodine), to proceedto some observation on the action of a physiologicor pharmacologic stimulus upon the renal circula-tion; or alternatively, to elevate the plasma levelof diodrast to 15 to 50 mgm. per cent of diodrastiodine, in order to saturate the tubules and meas-ure the " tubular excretory mass," i.e., the maxi-mal rate of tubular excretion (diodrast Tm).The initial clearance periods obtained on a par-ticular day and uncomplicated by any factor otherthan those specified we have averaged into a singledatum which we designate here as the diodrastclearance CD. Thus the data on CD recordedhere (Tables I and II) for a particular date repre-sent the average of 3 or more consecutive clear-ance periods. We feel that 3 such consecutiveperiods are the least upon which a trustworthyaverage figure can be based.2

DIODRAST Tm

For the measurement of diodrast Tm (TablesI, II and III), the plasma level of diodrast must

2 Chesley and Chesley (2) have misinterpreted ourpreliminary reports, in that they have suggested that wehave included in our standard observations clearance pe-riods when sodium sulphate was being administered.Sodium sulphate, or more recently mannitol, has beenused to maintain the urine flow in special circumstances,but such data are not included here.

739

W. GOLDRING, H. CHASIS, H. A. RANGES, AND H. W. SMITH

be raised to a point where the tubules are excret- (1) TD = UDV- PDCIrFWing this substance at a maximal rate. (Underthese conditions the diodrast clearance as ordi- GINF-W) PDCIN,narily defined (UVIP) has, of course, no direct /physiological significance.) The rate of tubular where UDand PD are the concentrations of dio-excretion of diodrast, TD, is calculated as the dif- drast iodine in mgm. per cc. of urine and plasma,ference between the total quantity excreted per respectively; V, the urine flow in cc. per minute;minute and the quantity excreted by filtration, i.e., CIN and CD the inulin and diodrast clearances in

TABLE I

Diodrast clearance, etc., in normal men

(Columns 5, 6, 7, 8 and 11 corrected to 1.73 sq. m.)

PlasmaSurface~ ~ ~ ~ ~ ceaane Effective Fltra- Rectal Diodrast

Subject Age Sare Date blood tion temper- | ntarea_Phenol|~ Diodrast flowfmnction ture

sq. m.

1.691.731.701.84

1.761.691.82

50 1.7838 1.72

1.70

1.861.62

1.61

1.70

1.63

1.83

1.38

1.671.631.781.781.651.56

2.081.62

April 9, 1937May 28, 1937May 28, 1937November 23, 1937November 29, 1937October 27, 1937November 17, 1937Ianuary 12, 1938anuary 28, 1938

February 18, 1938February 23, 1938March 7, 1938February 4, 1938March 28,1938April 8, 1938April 18, 1938April 25, 1938May 2, 1938March 30,1938May 11,1938April 22s 1938April 29, 1938May 6, 1938May 19, 1938April 6, 1938April 12,1938April 15, 1938April 13, 1938April 20, 1938April 28, 1938May 4,1938May 9, 1938May 13,1938May 23,1938May 18, 1938May 27, 1938May 16, 1938October 13, 1938November 8, 1938December 1, 1938December 2,1938October 24, 1938November 3, 1938November 18, 1938November 4, 1938October 25, 1938

cc.per

minute122128122144144110111156145150132141112101977782

100134111136136144139135109113178179161136117118121146140116163126145140112114988990

cc.per

minute

395375447429370294543469403418515293243165263257243390346349524494470320364395471526445412381369410532514363542397525436314355300351278

cc.per

minuk664778718956876650575993884720842986536444415409323471637680834955956852555604625818826687693831705727795841716878555885624498542565509529

cc.per

minut110713251177173816221205967

16231535133415591780891765692660521785

11431192160415901542135210051079107715251432118512001510134813201237134212851394925

15251095844874932955860

per cent

18.416.517.015.116.416.919.015.716.420.815.714.320.922.823.418.825.421.221.016.316.314.215.116.324.318.018.121.821.723.419.614.116.716.618.416.616.218.622.716.422.422.521.017.317.518.2

mgm.° F. iodixe per

minute

98.5100.5

98.5

39.051.255.0

98.5 1 45.1

99.0 1 57.6

98.2 43.098.2 72.0

98.6 1 58.3*

years

31

22

275028

A. M.

JA-.F. S.

T. S.J.W.V. C.

I. Ho.. Cw.

P.V.

P. M.P. F.

J. Cs.

C. J.

D. M.

B. L.

J. Ca.

S. M.J-.Ja.N.B.L. D.G. A.B. G.

D. K.M. S.

68

1832

36

24

49

27

43

352849394762

4826

I I I I II I I I I I I

740

RENAL FUNCTION IN NORMALMAN

TABLE i-Contwnued

PlasmaSurface cl~~~~~earmnces Effective Filtra- Reca Diodrast

Subject Age areaDate blood tion tempera- TMSurface Date |- flow fraction ture

Inulin Phenol red Diodmast ___I__ II

M. R.

W. G.R. D.

years

543028

sq. M.

1.732.061.98

F. A. 45 1.85

E. Mc.I 50

J. Co.W. S.P. Mc.

E. R.J.F.T-. W.

S. F.E. M.H. 0.

T. F.

H. S.

J. B.

W.J.

404854

394051312417

50

38

28

1.84

1.621.521.86

1.871.591.831.911.761.73

1.90

1.73

1.94

1.81

November 16, 1938November 23, 1938December 11, 1937January 3, 1938January 21, 1938January 31, 1938February 14,1938February 28, 1938December 12 1938December 28, 1938February 6, 1939February 24, 1939December 9, 1938December 29, 1938January 30, 1939February 8, 1939September 26, 1938October 18, 1938October 28, 1938November 1, 1938November 7, 1938November 14, 1938November 21, 1938November 28, 1938December 14, 1938January 4, 1939February 1, 1939December 27, 1938February 3, 1939November 17, 1939November 21, 1939November 29, 1939January 8, 1940January 5, 1940January 15, 1940January 19, 1940January 22, 1940January 24,1940January 29, 1940January 12, 1940January 26, 1940February 2, 1940January 31, 1940February 5,1940February 7, 1940February 21, 1940February 14, 1940

CC.

perminute

76131131125134139131131151155150199111

121113109125132134135119113117128114111

119140138

15313511610816614711812814991

10188

168160114125153

CC.

perminuts

199401452519550498472448542592

376

390359375395375345373341336

CC.

perminute

343780713719758768718684742843906

662846746743624600596663541601612579599510

496764647602730804582527716786687542680625449501762915738721778

CC.

perminute

746162211451110

1156117411061060116013801350

10501340126812941023953977

1087846974

1033958

'1010836800

12501085102712001312965868

134113811338890

12301042

778918

11771418122211741278

per cen

22.116.818.417.417.718.118.219.220.318.416.6

16.814.315.214.720.022.022.520.322.018.819.122.119.021.824.018.321.3

21.016.819.920.523.218.717.223.621.914.622.517.722.017.415.517.319.7

oF.mai.

iodisq perminute

99.2 1 56.8*

98.6 53.7

100.0100.9

98.798.899.398.3

98.7

98.5

99.0

98.6

45.1*61.0

54.447.234.658.1

49.7

38.8

62.4

56.9

Statistical summary entering each subject once onlyMean............................. 130t 391 688 1189 18.9 53.3ta ............................. 21.7 85.5 135.3 242.4 2.4 9.1

, .................................. 2.96 15.8 20.6 36.9 0.36 1.8n .............................. 54 29 43 43 43 26

Mean r Um i

Phenolred.3.1 0.34 0.063 29

Diodrast clearance 13.6 1.4 0.32 19Diodrast Tm.1. . .21

Inulin clearanceDiodrast Tm. | 2.57t 0.28 0.055 26

* Diodrast Tmis average of observations on this day with other observations in Table III.t Includes subjects shown in Table III.Ur, - Standard Error of Mean = a/n.

741

W. GOLDRING, H. CHASIS, H. A. RANGES, AND H. W. SMITH

TABLE II

Diodrast clearance, etc., in normal women(Columns 5, 6, 7 and 10 corrected to 1.73 sq.m.)

plainsclear- Ef- PU- Rec-

8ur- ances fec- tra- tal Dio-8bt Age face Date tive tion tem- drastject area blood frac- pera- TsInu- Dio- flow tion ture

lin drast

Pf iodineyears sq. m. m5 USR F. perns- '5- i- "'i min-

u ue ute steA. G. 24 1.36 October 14, 1938 92 445 718 20.7 98.5 36.6C. K. 30 1.67 November 8, 198 441* 726

November20, 1939 595* 948*November28, 1939 140 654 1027 214 98.5 63.8

D. W. 22 1.72 Decenber 1, 1939 145 722 1135 19.3 99.2 48.2December 8, 1939 782 1211

M. Z. 26 1.84 December 11, 1939 125 626 970 20.0December 18, 1939 113 584 863 19.4 99.4 46.5

E. R. 51 1.63 Dember 13, 1939 114 617 1094 18.5 98.8 42.7A. P. 39 1.52 December 15. 1939 109 554 938 19.7 99.4 40.7

December 22, 1939 102 538 871 19.0 99.3 39.3C. B. 49 1.57 December 20. 1939 101 550 789 18.4 99.2 43.4G. A. 30 1.46 December 26, 19839 147 708 1180 20.8C. 8. 29 1.70 December 27, 1939 115 500 875 23.0 99A 40.9R T. 29 1.69 January 3, 1940 122 666 1264 18.3 98.5 58.3L G. 26 1.57 January 17, 1940 105 559 954 18.8 98.8

Statistical summary entering each subject once onlyMean. 118.7 600.4 996 19.8 46.7a .................................. 17.5 87.0 162.7 1.4I 8.5. 11 11 11 11 9

Mean a n

Diodrast clearance 12.8 1.52 9Diodrast Tm

Inulin clearance 2.54 0.28 9Diodri t Tnm

*Omitted from average.

cc. of plasma per minute; F, the fraction of dio-drast which is free and therefore filtrable, as de-termined by the plasma concentration of dio-drast iodine and the plasma albumin content (seenomogram of Smith and Smith (13)); and Wistaken approximately as 1.00-per cent of plasmaprotein/100.

Albumin and total protein determinations were

made by us on 57 occasions on 31 subjects, and itwas found that FWaveraged 0.72 + 0.025. In a

subject with diodrast Tm of 52, inulin clearanceof 125 cc. and a plasma level of diodrast iodineof 20 mgm. per cent, a deviation in FWof + 0.02makes only + 1.0 per cent error in diodrast Tm.Wherever possible, however, protein determina-tion should be made for diodrast Tm measure-

ment in renal disease.In the determination of diodrast Tm, the total

"load " of diodrast carried to the tubules must,of course, exceed the maximal rate of excretion

if the tubules are to be saturated. The diodrastload/diodrast Tm ratio in our observations rangesfrom 1.5 to 5.0. Any ratio above 1.5 appears tobe satisfactory for this measurement.3 We rou-tinely take the average of 5 periods of ten to fif-teen minutes each for this datum.

It early became clear that diodrast Tm, involv-ing as it does a " saturation " phenomenon, ormaximal excretory activity on the part of the tu-bules, is very susceptible to changes in body tem-perature. Observations made upon the same sub-jects at different temperatures indicate a tempera-ture coefficient of about 2.0 for the excretoryprocess, i.e., a rise of 1.0° F. in body temperatureincreases the Tm value by about 10 per cent.Consequently, in all measurements of diodrast

TABLE III

Data on glucose Tm, etc., in normal men(Columns 6, 8 and 9 corrected to 1.73 sq.m.)

Inu- Rec-Sub- Sur- lin tal Glu- Dio-

jet Age face Date plasma tern- cose drast1eCt area clear- pera- Tm Tmance ture

cc. mgm. mgm.years sq. m. per 0 F. per per

minute minute minuteJ. J. 28 1.65 March 15, 1939 162 98.6 370 56.4J. Hu. 39 1.85 Feb. 13, 1939 128 98.4 322

Feb. 20, 1939 116 98.8 313 40.3W. S. 48 1.52 Feb. 10, 1939 148 99.0 316A. L. 42 1.88 Feb. 17, 1939 131 99.6 254R. W. 23 1.79 Feb. 23, 1939 149 98.7 323A. Lo. 41 1.95 April 12. 1939 157 99.0 380C. C. 42 1.77 May 15 1939 181 98.4 405 64.3

June 29,1939 134 98.5 349 63.7R. D. 28 1.98 May 24, 1939 150 98.0 336

June 9 1939 117 98.0 368 59.8June 15. 1939 117 97.8 51.8

H. M. 34 1.81 June 2,1939 150 97.8 461June 14 1939 138 98.2 440 65.7

J. B. 45 1.48 June 21, 1939 106 97.6 301 48.3June 28, 1939 127 97.7 331 59.7

P. M. 36 1.72 Oct. 11, 1939 93 97.8 243 55.3Oct. 16, 1939 119 97.8 264 62.4Oct. 23, 1939 114 97.8 275 59.4

T. S. 37 1.70 Oct. 18 1939 156 98.2 354 57.5Oct. 25, 1939 138 98.2 327 57.0

C. P. 34 1.67 Oct. 20 1939 152 99.0 397 58.2Oct. 27, 1939 161 98.2 433 56.7

E. Mc. 50 1.84 March 1. 1939 129 99.0 340 45.1Average 136 344 56.7

The load of diodrast carried to the tubules is theproduct of the plasma concentration and the renal plasmaflow. It is assumed that the renal plasma flow remainsthe same during diodrast Tm measurement as in the con-trol observations; hence we calculate the load as PD(CD-FWCIN), i.e., the actual plasma concentration of dio-drast during Tmmeasurement times the average diodrastclearance (CD) observed at low plasma diodrast levels,minus the quantity of diodrast filtered per minute. Fil-tered diodrast is deducted in this calculation inasmuch asafter diodrast has been filtered through the glomeruli itis no longer available for either tubular excretion or forsaturation of the tubules.

742

RENAL FUNCTION IN NORMALMAN

Tm (as indeed in all our observations) the rectaltemperature is recorded every twenty to thirtyminutes. The rectal temperatures observed by usrange from 97.5 to 99.5° F. and average 98.50F., and we have corrected all figures for diodrastTm reported here to 98.5° F. by adding or sub-tracting 10 per cent to the observed value for eachdegree of body temperature above or below thisvalue.

On a very few occasions we have observedtransient urticaria during diodrast Tm measure-ment. No serious disturbances have been en-countered, although we have on numerous occa-sions elevated the plasma level of diodrast to 50mgm. per cent of iodine or higher for periods ofone hour or more.

PHENOLRED CLEARANCE

The phenol red clearance was followed until De-cember, 1938, at which time it was abandoned infavor of other observations. Phenol red Tmhasbeen reported in only one subject (12) and is notrecommended.4

GLUCOSETmShannon and Fisher (9) have shown that, when

the plasma concentration of glucose is elevated toa sufficient level, the rate of reabsorption of glu-cose by the renal tubules in the dog becomesmaximal and constant. We here briefly reportthat we have found the same thing to be true innormal man, and include measurements of themaximal rate of tubular reabsorption of glucose,or glucose Tm,;in 14 subjects (Table III).

The rate of tubular reabsorption of glucose(TG) is calculated as the difference between thetotal quantity filtered per mninute and the quantityexcreted in the urine, i.e.,(2) TG = PGCIv - UGV,where CIN is the inulin clearance, PG and UGarethe concentrations of glucose per cc. of plasma

'Phenol red Tm measurement requires plasma con-centrations of 25 mgm. per cent, at which level the skindiscoloration is disturbing to the patient; moreover,phenol red, when given intravenously in large doses indogs, frequently causes nausea and a marked fall in thefiltration rate, and if used for this purpose in these orother animals should be given subcutaneously or intra-muscularly.

and urine, respectively, and V is the urine flowin cc. per minute. In order to raise TG to itsmaximal value (glucose Tm) (i.e., to saturate alltubules whose glomeruli are maximally or nearlymaximally active), the load of glucose deliveredto the tubules should exceed the maximal rate ofreabsorption by at least 20 per cent, i.e., the glu-cose load/glucose Tm ratio should exceed 1.20.The load of glucose carried to the tubules is theproduct of the plasma concentration and the rateof glomerular filtration in cc. of plasma, i.e.,PGCIN. At high plasma levels of glucose, slighterrors in the determination of plasma glucose orthe inulin clearance introduce disproportionatelylarge errors in the calculation of glucose Tm, andmaximal accuracy is attained at a load/Tm ratioof 1.4 to 1.6. Glucose Tmshould be determinedwith a rising or constant plasma glucose concen-tration, since the reabsorptive process exhibits anunexplained hysteresis (Shannon and Troast,personal communication).

The measurement of glucose Tm is valuablefor two purposes: First, it affords a measure ofthe quantity of intact tubular reabsorptive tissue,or the " tubular reabsorptive mass " of the kidney,in contrast to the " tubular excretory mass " asmeasured by diodrast Tm. Second, it affords amethod of evaluating the number of "active"nephrons, and hence open glomeruli, since at anappropriate plasma level of glucose a critical de-crease in filtration rate in a significant number ofglomeruli will cause the connected tubules to be-come unsaturated, and glucose Tmwill decreaseby a corresponding amount (7). The applicationof glucose Tm in appraising glomerular activitywill be discussed elsewhere; here we wish to re-port only the observed values under standardconditions.

The elevation of plasma glucose to the highlevels involved in this measurement (350 to 700mgm. per cent) may be expected to expand theplasma volume, under which conditions the renalblood flow might be modified either reflexly orthrough autonomous local adjustments. Evi-dence on this point will be presented separately;we need only note here that we do not considerthe data on the diodrast clearance when obtainedsimultaneously with the measurement of glucoseTmas physiologically comparable with the other

743

W. GOLDRING, H. CHASIS, H. A. RANGES, AND H. W. SMITH

data on this clearance presented here. Conse-quently, the diodrast clearance has been omittedfrom Table III. Wehave no reason, however,to believe that either the inulin clearance or thevalue of diodrast Tm is altered in normal subjectsunder these conditions, and we consider these data(Table III) as physiologically homogeneous withthose presented in Tables I and II and have ac-cordingly included them in the statistical summary.

THE QUESTION OF PHYSIOLOGICALLY ACCEPTABLE

iSTANDARDp VALUESOF THE DIO-

DRAST CLEARANCE5

Inulin, phsnol red and diodrastThe first point to be considered in this connec-

tion is the possibility of hyperemia induced bypyrogenic inulin. The presence of pyrogen ininulin and a method of removing the pyrogen werereported by Goldring and Smith in 1936 (5), anda more effective method of purification was re-ported by Smith, Chasis and Ranges in 1938 (11).In adequate doses this pyrogen can produce renalhyperemia (10). Ever since the hyperemic prop-erties of this pyrogen were discovered, we havebeen aware of the danger of this action compli-cating our observations. That such has not beenthe case we are convinced for the following rea-sons: When pyrogenic inulin is given in eventhreshold doses, the first reaction is a rise in tem-perature two to four hours after administration,usually with complaints of chilliness and othersubjective reactions. We have never succeededin producing renal hyperemia by doses too smallto produce the febrile reaction. When adminis-tered in larger doses, there has, in our experience,invariably been a latent period of at least ninetyminutes before the development of renal hyper-emia. Since 1936 rectal temperatures have beentaken routinely on all subjects, both during ob-servation and every hour thereafter until 4:00p.m., and in all the data on diodrast clearance re-corded here, the observations have been concludedwithin approximately sixty minutes after the firstinulin injection. In no instance has the afternoon

5 The question of quantitative correspondence betweenthe plasma diodrast clearance and the actual renal plasmaflow (or between the whole blood diodrast clearance andrenal blood flow) will not be discussed here since it willbe treated in a forthcoming publication.

temperature exceeded 1000 F., nor have therebeen subjective evidences of pyrogenic reaction.Consequently, we set aside as improbable the pos-sibility that our mean value for the diodrast clear-ance has been elevated by pyrogenic hyperemia.

With regard to diodrast, it is known that therapid intravenous administration of relativelylarge doses of this substance cause a transitoryfall in blood pressure (3), but the doses involvedexceed many times the quantities which we usein clearance determinations. These quantities aretypically 350 mgm. (1 cc.) as a priming dose, fol-lowed by infusion at the rate of 16 mgm. perminute. Smith, Goldring and Chasis (12) foundthat small doses of diodrast (0.35 to 1.5 grams)do not increase the phenol red clearance, and morerecently we have made numerous observations inwhich the plasma concentration of diodrast hasbeen slowly elevated from 1.0 to 10 mgm. percent, without any increase in diodrast clearance.On the basis of this evidence we exclude thesupposition that diodrast has produced renalhyperemia.ff

With regard to phenol red, Herrick, Mann andSheehan (6) have reported that injections of thissubstance increase the renal blood flow (thermo-stromuhr method) in unanesthetized dogs. Theaverage dose used by these investigators was 15cc. of a 6 per cent solution injected in one minute,or 900 mgm. per minute. In one of the two ex-periments illustrated by them the disturbance inrenal blood flow had passed off in ten minutes,and in the other there was practically no change.In both experiments the dog vomited immediatelyafter the injection. The largest quantity ofphenol red we have used has been 30 mgm. perminute for ten minutes as a priming dose com-pleted twenty minutes before the first clearanceperiod, followed by infusion at 4 mgm. per min-

's This applies only to the measurement of the diodrastclearance at low plasma levels of diodrast. Followingthe injection of 30 to 40 cc. of diodrast for the measure-ment of diodrast Tm, there appear to be occasional dis-turbances in the renal blood flow, as indicated by a slightto moderate fall in the inulin clearance. The circulatoryeffects of these large doses cannot be examined by thediodrast clearance method, since the tubules are then satu-rated and the extraction ratio depressed. But any circu-latory effects of large doses given relatively quicklywould not controvert the evidence that small doses haveno such effect.

744

RENAL FUNCTION IN NORMALMAN

ute (12); assuming a 10 kgm. dog and 60 kgm.man, the dose given by Herrick, Mann and Shee-han is about 180 times our priming infusion, and1350 times our maintenance infusion. In the ab-sence of better evidence, we therefore discountthe possibility of phenol red having induced renalhyperemia.

PSYCHOGENICDISTURBANCE, SELECTION, AND

HYDRATION

In brief, we believe that the chemical elementsinvolved in clearance determinations have intro-duced no significant error in the data of Tables Iand II. Weare, however, aware of three factorswhich should be considered in presenting our dataas " standard" normal values.

First, there is the factor of psychogenic disturb-ance. The mere procedures of catheterization,venipuncture, and the attendance of 3 or morepersons are obvious sources of apprehension tosome patients. The actuality of psychogenicvasoconstriction in the kidney has recently beendemonstrated (10) and, though the question isnot open to experimental investigation at the pres-ent time, we fully recognize the possibility thatsuch apprehension as may unavoidably be associ-ated with the procedure of clearance determina-tion may produce renal vasomotor disturbance.Where our subjects have been examined on 2 ormore occasions, the diodrast clearance has beenlarger in the second than in the first examinationless than 50 per cent of the time (Tables I andII). Nevertheless, nearly all our data involverepeated observations on " trained " subjects, i.e.,subjects who are quite accustomed to clearanceexamination, and this fact may have led to a figurefor the mean diodrast clearance somewhat abovethe figures which have been reported by others(2, 17).

Secondly, we recognize the possibility that selec-tion may have influenced our data. Many of oursubjects have been selected with a view to theexamination of pharmacologic agents, etc., and itmay be that we have selected a group of subjectswhich, in respect to renal function, is not fairlyrepresentative of the general population. Wewish merely to call attention to this possibility forwe strongly discount its actuality.

The third factor bearing on this point is the

degree of hydration of the subject. Our sub-jects take a liter of water the night before, andanother liter on the morning of the examination(the last water at least ninety minutes before thefirst clearance period) and, although we have ob-served no difference in a few comparative obser-vations on the same subjects, it may be that thisthorough hydration results in an increased renalblood flow. But even if such is the case, we be-lieve that a standard technique such as this ispreferable to one which will permit variable de-hydration to enter as a complication.7

In view of the above considerations, we presentour data as "standard" rather than "basal-" 8observations, emphasizing that the conditions un-der which they have been made are both physio-logical and reproducible. The technique of mak-ing all such observations must be improved asknowledge of the renal circulation is increased.

DISCUSSION

The larger series of data presented in Table Irequires no discussion other than to note that theaverage inulin and phenol red clearances comparefavorably with the average values previously re-ported for 25 subjects (12). The present aver-age diodrast clearance is substantially less thanthat previously reported, but the previous figurewas based on only 6 subjects, 2 of whom werecounted twice.

The most significant aspect of the present datais the quantitative relationship which exists be-tween the diodrast clearance and the inulin clear-ance, on the one hand, and diodrast Tm, on theother. Examination of the data on those sub-jects in whom diodrast Tm has been measuredshows that this term has a close positive correla-tion with both clearances, as shown in Figures 1and 2.

7 Werecommend thorough hydration in order to main-tain the urine flow and to permit accurate clearance pe-riods. Where both hydration and Na2SO, are contra-indicated, 7 to 8 per cent mannitol or sorbitol (Abbottampules) may be incorporated in the infusion as a diu-retic. Dr. Catherine Welsh and her co-workers (per-sonal communication) have used the hexitols for thispurpose in clearance studies on normal pregnant womenand women with pre-eclampsia.

8 The ambiguity of the term " basal " is illustrated bythe recent studies of Scott, Bazett and MacKie (8) onthe circulation under various climatic conditions.

745

W. GOLDRING, H. CHASIS, H. A. RANGES, AND H. W. SMITH

TABLE IV

Statistical analysis of subjects with diodrast Tm(Men and womenare treated as one series)

Numberof Range Mean u mean

subjects xioo

Diodrast whole blood clearance, cc. per 1.73 sq. m. per minute .. . 28 718-1566 1115 219 19.6Diodrast plasma clearance, cc. per 1.73 sq. m. per minute ..... .... 28 455-921 669 128 19.1Inulin plasma clearance, cc. per 1.73 sq. m. per minute ....... 35 92-179 131 22.1 16.9Diodrast Tm, mgm. iodine per 1.73 sq. m. per minute ... 35 36.6-72.0 51.6 9.4 18.2Diodrast whole blood clearance

Diodrast Tm..clearance 28 16.0-29.5 22.3 2.70 0.77 12.1Diodrast plasma clearance

Diodrast T.nm.28 10.2-16.7 13.4 1.4 0.74 10.5Inulin clearance 35 2.09-3.12 2.56 0.28 0.77 9.0

__xy|2(m - x)' -mzm7,

\I() -n r =- *l * =vThe value of r in column 6 refers to the correlation between the numerator and denominator.

DIODRAST TM MGM. IODINE PER 1.73 SQ. M. PER MIN.

FIG. 1. THE PLASMA DIODRAST CLEARANCE(EFFmvE RENAL PLASMAFLOW) IN 28 SUBJECTS (MEN ANDWOMEN)IN RELATION TO DIODRAST Tm (TUBuLAR EXCRETORYMASS), AS REcoRDEDIN TABLES I AND II

Each subject enters once only into the calculation of the correlation coefficient (0.77). Diodrast Tm may betaken as proportional to the total functional tubular tissue, and therefore closely proportional to the kidney weight.

746

RENAL FUNCTION IN NORMALMAN

a

eeO A'i | 0 ^

ao a

.

) .

15 20 25 30 35 .40 45 50 55 aDIODRAST TM MGM. IODINE PER 1.73 SQ. M. PER MIN.

65 - 70

FIG. 2. THEPLASMAINULIN CLEARANCE(FILTATION RAm) IN 35 SUBJECTS (MEN AND WOMEN)IN RELATIONTro DIODAST Tm (TuBuLR EXCRETORYMASS), AS RECORDEDIN TABLES I, II, AND III

The correlation coefficient is 0.77.

In Table IV we have calculated the coefficientof correlation between diodrast Tmand the wholeblood diodrast clearance, the plasma diodrastclearance and the inulin clearance. There are alsopresented in this table the standard deviations forthese clearances, calculated on the same subjects.(Each subject is here considered once only.Data are available on 35 subjects for the inulinclearance, but on only 28 subjects for diodrastclearances since on 7 subjects diodrast Tm was

determined simultaneously with glucose Tm, un-

der which conditions, as we have pointed outabove, the diodrast clearances are physiologicallysuspect.)

Reference to Table IV shows that the standarddeviation expressed in per cent of the mean issubstantially smaller when the clearances are re-

lated to diodrast Tm than when they are consid-ered independently. Weinterpret this fact to in-dicate that differences in the clearances are in partattributable to such differences in the size of thekidneys as are not excluded by the usual correc-

tion for body surface area, these differences inkidney size being evident in data on normal kidneyweights (16). To what extent diodrast Tm isstrictly proportional to kidney weight is at presentunknown, but since diodrast Tm is in theory a

measure of the total quantity of tubular excretorytissue in the kidney, we would expect the corre-

lation between these two terms to be of a signifi-cantly high order. In this view, diodrast Tmmaybe taken as a measurement of kidney size, andfrom the close correlations shown in Table IV andFigures 1 and 2, we would be permitted to say

that the whole blood or plasma diodrast clearanceand the inulin clearance in various individuals are

relatively more constant when compared per unitof kidney weight than when compared per sq. m.

of body surface area.

The present series of data on glucose Tm is toosmall to afford an accurate statistical analysis.The average glucose Tm/diodrast Tm ratio in 10subjects is 6.49.

Relative to the data in Table IV, it should be

200CcL&J0.180

6 160U)

F 140ar

-12VLJI

Li', 00uz

cc 804LJU 60z-

z 40

747

W. GOLDRING, H. CHASIS, H. A. RANGES, AND H. W. SMITH

noted that according to statistical theory it is im-possible to define the absolute range of " normal "values. Statistical analysis at best affords a state-ment of the probabilities that a given observationbelongs to the population which has been analyzed.If in this population there is a symmetrical dis-tribution of the individuals about the mean, then95 per cent of the individuals will fall withinm -+- 2a, and 99.5 per 'cent within 3a. Or, con-versely, there is one chance in 20 that an indi-vidual actually belonging to this population mayfall outside the range of + 2a, and one chance in200 that it may fall outside the range of + 3a.If, for example, a particular diodrast clearancefalls just below 413 or just above 925 (+2a),there is still one chance in 200 that it is " nor-mal " with respect to the data of Table IV. Theabsolute identification of this subject as havingabnormal renal function is statistically impossible.

Thirty-five subjects is, moreover, so small a sta-tistical series that it affords a fair chance of sam-pling error. Wewould, at the present time, placemore emphasis upon the physiological sources ofvariation discussed above than upon the statisticalsignificance of this limited series.

Wecan briefly record here that we have utilizedthe ballistocardiographic method of Starr, et al.(15) for the measurement of cardiac output be-fore and during the clearance determination in 6subjects (12 observations). In no case have wefound an increase in cardiac output, as referredto a control measurement on the morning of ob-servation, to be associated with the clearance pro-cedure (infusion, catheterization, etc.). This in-vestigation is being extended by other methods ofmeasuring cardiac output in order to confirm theaccuracy of the ballistocardiographic method inour hands.

SUMMARY

Data on the diodrast and inulin clearances, andon the maximal rate of tubular excretion of dio-drast (diodrast Tm) and the maximal rate oftubular reabsorption of glucose (glucose Tm) are

presented for a series of normal men and women.

Diodrast Tn, a measure of the total quantityof tubular excretory tissue in the kidney, is con-

ceived to be roughly proportional to the totalquantity of renal parenchyma.

Statistical analysis of the data reveals that boththe diodrast and inulin clearances increase in pro-portion to diodrast Tm, and that comparison ofclearances in different individuals is more accuratewhen made on the basis of diodrast Tmthan whenthe absolute values of the clearances are consid-ered alone. This method of comparison is func-tionally equivalent to expressing the renal plasmaor whole blood flow and the filtration rate on thebasis of unit mass of renal parenchyma.

The significance of the mean values of the dio-drast clearance, etc., is discussed in relation to thepossible perturbation of renal blood flow by fac-tors involved in the clearance procedure, such asinulin and diodrast, and by selection of subjects,hydration and psychic influences. It is believedthat the data accurately reflect the renal bloodflow under " standard" conditions which closelyapproach the ordinarily accepted concept of basalconditions in the systemic and renal circulation.

The authors are indebted for unfailing cooperation toDr. Willie W. Smith, Katharine S. Tilson, Helen A.Keigher, Anna Rosenthal, Norma Finkelstein, FrancesE. Marx, Anna M. Lublin, Betty J. Crawford, MarthaJ. Barrett and Lucy M. Aliminosa, who have at onetime or another participated as analysts in this investiga-tion; to Barbara A. Parker and Mr. Sylvan M. Stillman,who have assisted in the measurement of the ballisto-cardiographic records; and to Nurse Ann S. Rivoire.

METHODS

Only important changes from methods previously de-scribed (12) are detailed below:

1. Infusion fluid. This is prepared in 0.9 per centsaline (Sterisol saline, Schering and Glatz, Inc.) andadministered from an open infusion flask without pre-cautions for maintaining temperature, the rate beingtimed by means of a Murphy drip chamber and regulatedby means of a 4-inch tunnel clamp. Typical protocolsfollow:

Diodrast clearance. To 800 cc. of saline are added 12cc. of 35 per cent diodrast solution (Winthrop ChemicalCompany) and 25 cc. of 10 per cent inulin solution (U. S.Standard Products Company). Immediately after theinfusion is started (zero minutes), 1.3 cc. of diodrastsolution and 15 cc. of inulin solution are injected intothe infusion tubing as a priming dose. (A control bloodsample (B.) is drawn when venipuncture is first made.)U. is collected at 20', B1 at 22', U, at 32', U, at 44', U.at 56', and B, at 54' or 58'.

Diodrast Tm. The infusion tubing is disconnectedfrom the needle, the infusion fluid drained to 400 cc.;11.5 cc. of diodrast are added to each 100 cc. of infusion

748

RENAL FUNCTION IN NORMALMAN

left in the flask and mixed, and 26 cc. of diodrast areinjected into the infusion tubing as a priming dose. Theinfusion is accelerated for a short time to empty thedead space of dilute diodrast, the rate then being restoredto 4 cc. per minute. Twenty minutes later (ca. 79'),U. (second discard) is collected, B. is drawn at 81', U4at 91', U. at 103', B4 at 113', U, at 115', U7 at 127', Usat 139', and B, at 137' or 139'.

In the average normal subject the above infusion willgive 10 mgm. per cent of inulin in the plasma, 1.5 mgm.per cent of diodrast iodine between B1 and B,, and 30(mgm. per cent between B. and B,. If the clearances areapproximately known, the appropriate quantities of inulinand diodrast can be related to the desired plasma iodineconcentration in mgm. per cent (PD) and body weight-(kgm.) on the basis that the priming dose is distributedin approximately 20 per cent of the body weight, andthat at a constant rate of infusion, this rate will equalUV. That is, the priming injection in cc. should equal

P7kgm. cc., and 700D cc. of 35 per cent diodrast solu-

tion should be added to each 100 cc. of infusion. ForTmD, the priming injection is calculated in the samemanner and the sustaining infusion should containTMD+ PD. c2TMD cC. of diodrast solution, since UV

= TmD+ PDFW2.57 TmD/100. With 10 per cent inu-

lin solution the priming dose should be ' kg cc. and

PINCINeach 100 cc. of infusion should contain 400 cc.Glucose Tm. Glucose (50 per cent solution) to the

extent of 18 per cent is incorporated in the infusion fluidand 25 grams of glucose are injected into the infusiontubing as a priming dose, and 25 grams of glucose (pure)are given per os at zero minutes and again about midwayin the observations. Depending on glucose Tm, thisyields 350 to 750 mgm. per cent glucose in arterial blood.If the hand is warmed in hot water, the venous bloodfrom the veins of the hand or wrist may be used insteadof arterial blood, there being no A-V difference at thislevel of hyperglycemia and constant blood glucose level.The blood glucose level should be constant or rising forthis measurement. It has been our experience that sub-jects showed marked differences in capacity to store andutilize glucose. Since glucose storage and utilizationseem usually to be accelerated after the administrationof a glucose infusion for an hour or so, it is most dif-ficult to maintain the requisite plasma level of glucoseduring the second hour.

2. Analytical methods. We have recently adopted theinulin method of Alving, Rubin and Miller (1), with themodification recommended by them (personal communi-cation). It has been our experience that pure diphenyl-amine is necessary in this method. We have found thematerial supplied by Pfanstiehl Chemical Company andthe G. Frederick Smith Chemical Company to be satis-factory. In order to conserve blood we have changedthe technique slightly from the original description:

Glucose is removed from blood and urine by treat-ment with a yeast suspension which has been well washed(6 to 8 times). This suspension will keep for a weekor longer in the icebox, but must be centrifuged andfreshly suspended on the day of use. Two cc. of plasmaare mixed with 6 cc. of ca. 20 per cent yeast suspension(of known hematocrit) in 16 X 125 mm. pyrex tubes.The mixture is occasionally agitated, and centrifugedafter fifteen minutes at room temperature. From 2 to 5cc. (typically 2 cc.) of the supernatant fluid are removedto a 50-cc. Erlenmeyer flask containing 5 cc. of waterand 6 cc. of ZnSO4-H2SO4 mixture prepared as describedby Somogyi (14). This mixture is well agitated and2 cc. of 0.75 N NaOH are added. The flask is closedwith the finger and well shaken. The mixture is centri-fuged in 20 X 150 mm. heavy-walled pyrex tubes andfiltered through a pledget of washed cotton. The totaldilution should yield about 0.50 mgm. per cent of inulinin the filtrate. Five cc. of the above filtrate are trans-ferred in duplicate to the special tube recommended byAlving, et al., and the inulin determination carried outwith the diphenylamine reagent according to their speci-fications. Color is read in an Evelyn colorimeter, usinga 635 filter. The urine samples, after dilution to a U/Pratio of approximately 1.0, are yeasted, precipitated andanalyzed in the same manner as the plasma.

For the determination of the yeast plus plasma inu-linoid blank, 2 cc. of plasma from B, plus 2 cc. of 5 mgm.per cent inulin are treated with 6 cc. of yeast, and 3 cc.are precipitated as above. This blank varies from 0.1 to0.7, and usually falls in the lower range of these figures.Duplicate inulin standards containing 02, 0.5 and 0.8mgm. per cent, as well as triplicate reagent blanks areincluded. Apart from preparing washed yeast and wash-ing apparatus, one analyst, when using an 8-tube, I. E. C.centrifuge, can complete duplicate determinations on 5plasma and 8 urine samples, with one B. sample, in eighthours.

Glucose in plasma and urine is analyzed by the Folin(4) method, using the technique of a single standard asdescribed by Smith, et al. (12).

Iodine is analyzed by the Kendall method as modifiedby Smith, et al. (12), except that 3 cc. instead of 5 cc.of plasma are now used for low plasma iodine levels (1to 2 mgm. per cent). Two cc. of plasma are used forhigh plasma iodine levels. An automatic filling 2-cc.burette, graduated in 0.01 cc. (E. Machlett and Sons,New York) placed in a well-lighted cabinet with white-interior, has been found extremely useful for titration.Six plasma and 12 urine samples can easily be analyzedin duplicate in eight hours by one analyst. This methodhas repeatedly been demonstrated to give 100 per centrecovery of added iodine in organic form and yields verysatisfactory checks. The chief difficulty encountered inits use has been in the growth of mold on the inner wallsof the 5-gallon bottles which we commonly use for dis-tilled water. This mold becomes detached and catalyzesthe oxidation of KI to I.. The difficulty has been cir-

749

W. GOLDRING, H. CHASIS, H. A. RANGES, AND H. W. SMITH

cumvented by washing the bottles with cleaning fluidonce every week or so.

For the determination of phenol red, 2 cc. of plasmaare added to an S-tube (Evelyn colorimeter) containing10 cc. of 0.9 per cent NaCi. While following the opticaltransmission on the galvanometer (540 filter), the mix-ture is acidified with 5 per cent acetic acid (3 to 5 drops)until maximal transmission is obtained (just short ofprecipitation of proteins) and used to set the galvanom-eter at 100. Two drops of saturated Na.CO, are thenadded, the contents of the tube mixed, and the transmis-sion read on the galvanometer scale. The phenol redcontent is determined from a standard curve preparedin this manner with plasma and known concentrationsof phenol red.

BIBLIOGRAPHY

1. Alving, A. S., Rubin, J., and Miller, B. F., A directcolorimetric method for the determination of inulinin blood and urine. J. Biol. Chem., 1939, 127, 609.

2. Chesley, L. C., and Chesley, E. R., The diodrast clear-ance and renal blood flow in normal pregnant andnon-pregnant women. Am. J. Physiol., 1939, 127,731.

3. Edwards, E. A., and Biguria, F., A comparison ofskiodan and diodrast as vasographic media, withspecial reference to their effect on blood pressure.New England J. Med., 1934, 211, 589.

4. Folin, O., Two revised copper methods for bloodsugar determinations. J. Biol. Chem., 1929, 82, 83.

5. Goldring, W., and Smith, H. W., Inulin and its suit-ability for intravenous administration in man.Proc. Soc. Exper. Biol. & Med., 1936, 34, 67.

6. Herrick, J. F., Mann, F. C., and Sheehan, H. L., Theinfluence of phenol red and creatinine on the renalblood flow. J. Pharmacol. & Exper. Therap.,1939, 66, 73.

7. Ranges, H. A., Chasis, H., Goldring, W., and Smith,H. W., The functional measurement of the numberof active glomeruli and tubules in the kidneys ofnormal and hypertensive subjects. Am. J. P-hysiol.(Proc.), 1939, 126, 603.

8. Scott, J. C., Bazett, H. C., and MacKie, G. C., Cli-matic effects on cardiac output and the circulationin man. Am. J. Physiol., 1940, 129, 102.

9. Shannon, J. A., and Fisher, S., The renal tubular re-absorption of glucose in the normal dog. Am. J.Physiol., 1938, 122, 765.

10. Smith, H. W., The physiology of the renal circula-tion. Harvey Lect., 1939-40. (In press.)

C-11. Smith, H. W., Chasis, H., and Ranges, H. A., Suit-ability of inulin for intravenous administration toman. Proc. Soc. Exper. Biol. & Med., 1938, 37,726.

_ 12. Smith, H. W., Goldring, W., and Chasis, H., Themeasurement of the tubular excretory mass, ef-fective blood flow and filtration rate in the normalhuman kidney. J. Clin. Invest., 1938, 17, 263.

13. Smith, W. W., and Smith, H. W., Protein binding ofphenol red, diodrast, and other substances inplasma. J. Biol. Chem., 1938, 124, 107.

14. Somogyi, M., A method for the preparation of bloodfiltrates for the determination of sugar. J. Biol.Chem., 1930, 86, 655.

15. Starr, I., Rawson, A. J., Schroeder, H. A., and Jo-seph, N. R., Studies on the estimation of cardiacoutput in man, and of abnormalities in cardiacfunction, from the heart's recoil and the blood'simpacts; the ballistocardiogram. Am. J. Physiol.,1939, 127, 1.

16. Wald, H., The weight of normal adult human kidneysand its variability. Arch. Path., 1937, 23, 493.

17. White, H. L., Findlay, T., Jr., and Edwards, J. C.,Interpretation of diodrast clearances in man.Proc. Soc. Exper. Biol. & Med., 1940, 43, 11.

750