sodium homeostasis in chronic renal disease · sodium homeostasis in chronic renal disease...

Kidney International, Vol. 21(1982), pp. 886—897

NEPHROLOGY FORUM

Sodium homeostasis in chronic renal diseasePrincipal discussant: NEAL S. BRICKER

University of California School of Medicine, Los Angeles, California

Case presentation

A 63-year-old man was admitted to the Wadsworth Veterans Admin-istration Hospital for evaluation of azotemia and hypertension. Thepatient had been in apparent good health until one year earlier, whenstudies during a hospitalization elsewhere revealed a serum creatinineof 5.2 mgldl, a BUN of 45 mg/dl, and 3+ proteinuria. An intravenouspyelogram was within normal limits. Hypertension was noted onadmission and antihypertensive medications were given. The patientsubsequently was lost to follow up. Forty years previously the patienthad been hospitalized because of transient gross hematuria of unknowncause.

On admission to Wadsworth, the patient's blood pressure was 170/100mm Hg and funduscopic examination revealed artenolar narrowing. Thecardiopulmonary and abdominal examinations were normal, the prostatewas not enlarged, and no edema was present. The rest of the physicalexamination was within normal limits. Laboratory findings revealed:serum creatinine, 10.6 mgldl; BUN, 81 mg/dl; serum sodium, 139mEq/liter; potassium, 5.3 mEq/liter; chloride, 107 mEg/liter; bicarbonate,16 mM/liter; calcium, 7.2 mgldl; and phosphorus, 4.5 mgldl. The serumalbumin, bilirubin, glucose, uric acid, and transaminase levels all werenormal. Ultrasound examination revealed small kidneys bilaterally.

The patient was given a diet containing 46 mEq of sodium as part ofhis antihypertensive regimen. Within 36 hours he became progressivelysomnolent. The blood pressure fell to 90/60 mm Hg in the recumbentposition and 70 mm Hg when the patient was standing.

The patient was transferred to the Metabolic Ward, where 2 liters ofisotonic saline were administered and a dietary sodium intake of 150mEq/day was instituted. Following stabilization on this diet, the pa-

tient's blood pressure returned to 150/100 mm Hg, and he requiredpropranolol for optimal blood pressure control.

As part of a protocol to determine the functional adaptation ofsurviving nephrons in advanced chronic renal failure, the patient wassubjected to a water immersion study (see Fig. 1). Control plasmavolume, measured by dilution of radiolabeled albumin, was 1908 ml andthe body weight was 71.4 kg. The patient was immersed for 4 hours inthe seated position in a tank of water (kept at a constant temperature of354 5 C) up to his neck. Both sodium excretion and fractionalexcretion of sodium (FENa%) increased during immersion and returnedto nearly baseline levels during the recovery period. Sodium excretionincreased from 113.3 iEq/min in the 60-minute pre study period to 252iEq/min during the third hour of immersion and returned to 101.3iEq/min in the recovery (postimmersion) period. The FENa% rose from9.1% in the prestudy period to 18.8% during the third hour of immersionand returned to 10.1% in the recovery period. Plasma renin activitydecreased from 1.7 ng/ml/hr in the prestudy period to 0.62 ng/ml/hr bythe fourth hour of immersion. The inulin clearance remained stablethroughout the study at 9 mI/mm. Following this study the sodiumcontent of the patient's diet was reduced by 20 mEq per day at weeklyintervals.

By the second week of the study, the patient was noted to be losingmore sodium than was provided in the diet. He became hypotensive andagain required intravenous saline to restore a euvolemic state. Dietarysodium content was returned temporarily to a higher level and subse-quent decrements in sodium intake were carried out more gradually.After 3 weeks it was possible to discontinue the antihypertensivemedication and, after 12 weeks, dietary sodium intake had beenreduced to 15 mEq/day. At the end of this period, the patient's weightwas 68 kg and his plasma volume was 2174 ml. He was excreting 13 to20 mEq of sodium daily.

A second water immersion study revealed that sodium excretion andFENa% again increased during the period of immersion, but bothcontrol and immersion values were considerably lower than thoseobserved during the initial study. Sodium excretion increased from 4Eq/min in the prestudy period to 55 p.Eq/min by the third hour andreturned to 29.6 i.Eq/min in the recovery period. Similarly, the FENa%increased from 1.49% to 6.9% by the third hour and returned to 3.1% inthe recovery period. The plasma renin activity decreased from 3.2nglml/hr to 1.9 ng/ml/hr during the fourth hour of immersion. Measure-ment of natriuretic factor by rat bioassay, which had been negativeduring the first water immersion study, turned positive during thesecond. The inulin clearance was lower than before (6 mI/mm) butremained stable during the study.

Following completion of this study, the patient was given a dietcontaining 80 mEq/day of sodium and was observed for 7 days, afterwhich vascular access was created for subsequent dialysis. On dis-charge the patient had a serum creatinine of 10.8 mg/dl and a bloodpressure of 130/85 mm Hg.

Presentation of the Forum is made possible by grants from Smith Kline& French Laboratories, CIBA Pharmaceutical Company, GEIGYPharmaceuticals, and Boehringer Ingelheim Ltd.

0085-2538/82/0021-0886 $02.40© 1982 by the International Society of Nephrology

DiscussionDR. NEAL S. BRICKER (Professor of Medicine, Director,

Program in Kidney Diseases, UCLA School of Medicine, LosAngeles, California): If one could eliminate nephrons one at atime in 2 million consecutive steps while maintaining salt intakeconstant, the continued preservation of sodium balance would

886

EditorsJORDAN J. COHENJOHN T. HARRINGTONJEROME P. KASSIRER

Managing EditorCHERYL J. ZUSMAN

Michael Reese Hospital and Medical CenterUniversity of Chicago Pritzker School of Medicine

andNew England Medical Center

Tufts University School of Medicine

Nephrology Forum 887

2 3 4 5

o l5OmEqNadiet.15 mEq Na diet• Immersion commenced• Immersion discontinued

Fig. 1. Effect of water immersion on fractional excretion of sodiumduring two levels of dietary sodium intake. Upper curve representsresults of water immersion to neck in "adapted" uremic patient. Lowercurve represents results of water immersion in same patient following"de-adaptation." Details of de-adaptation are included in text.

require resetting the rate of sodium excretion by all survivingnephrons 2 million times. Thus, each time another nephron isdestroyed, the average rate of sodium excretion by eachsurviving nephron would have to increase to avoid permanentsodium retention, even though the mean increment might be assmall as 5 x 1O mEq/min/nephron.

Although single-file destruction of nephrons does not occur inchronic renal disease, it is the rule rather than the exceptionthat sodium balance is maintained until the vast majority of theoriginal nephron population has been destroyed. How a systemmight operate to preserve sodium homeostasis in the presenceof the relentless destruction of nephrons and a constant saltintake, effecting precise changes in the average rate of sodiumexcretion per nephron as small as those quoted above, consti-tutes a major area of current interest and inquiry.

Special problems created by chronic renal disease

Variation in excretory rates among nephrons. Figure 2 de-picts values for single-nephron sodium excretion rates in 3hypothetical subjects ingesting the same amount of salt daily:(1) a normal person; (2) a patient with chronic renal diseasesecondary to chronic glomerulonephritis; and (3) a patient withchronic interstitial nephritis. The three bars for each subjectrefer to different subsets of nephrons, based on mean values forsingle-nephron GFR (SNGFR). The data, as will be discussed,provide a compelling case for the existence of a complexbiologic control system for the regulation of sodium homeosta-sis. I will attempt to describe the biologic task confronting sucha control system and to emphasize some of the obstacles thatmust be overcome if the task is to be accomplished. Finally, I

• I] 3! 60Th0SNGFR

Fig. 2. Sodium excretion per nephron (nEqimin) when the intake is 120mEqlday. Each bar represents one-third of the total nephron populationfor each of three hypothetical subjects based on mean values for single-nephron GFR. The values for single-nephron GFR for each of thesubgroups are shown in Table 2. The basis for the calculations used toderive the values for sodium excretion rate per nephron is described inthe text.

will present a model for a control system that can operate inchronic renal disease as well as in health.

The data in Figure 2 are based on the following assumptions:All 3 subjects are the same size (70 kg), have the sameextracellular fluid volume (14 liters), and have the same serumsodium concentration (140 mEq/liter). All 3 ingest the sameamount of sodium daily (120 mEq/day), and all 3 excrete thesame amount of sodium daily (120 mEq); hence all 3 maintainexternal sodium balance with equal precision.

Table 1 presents an estimate of the number of nephrons ineach subject and of the mean value for SNGFR of thesenephrons. The normal person has a GFR of 120 mI/mm and 2million nephrons. Both patients have GFRs of 2 mI/mm. How-ever, patient A, who has chronic glomerulonephritis, has100,000 nephrons, whereas patient B, who has interstitialnephritis, has approximately 33,000 nephrons. The reason forthe threefold difference in the number of nephrons in 2 patientswith equal whole-kidney GFRs is that the average value forsingle-nephron GFR is three times as great in patient B as inpatient A.

In each subject, as already noted, the nephrons have beendivided into three subgroups, each having a different meanvalue for SNGFR. These data are shown in Table 2. To simplifythe calculations, each subgroup is assumed to have one-third ofthe total number of functioning nephrons. I should point outhere that the values assigned for SNGFR represent extrapola-tions from studies in experimental animals with normal kidneys,kidneys with severe glomerular lesions, and kidneys withinterstitial lesions [1—121.

Figure 2 clearly shows that the values for sodium excretion

Chronic GNGFR 2 mI/mm

100,000 nephrons

zuJU-

SNGFR 2 mI/mm

33,400 nephrons3.8

3.4

3.0

2.6

c 2.2Ea.

L 1.8

1.4

1.0

0.6

0.2

NormalGFR 120 mI/mm

2,000,000 nephrons

50 60 70

Hours

888 Sodium homeostasis in chronic renal disease

Table 1. Details of adaptation in sodium excretion: Total number ofnephrons and mean SNGFR valuea

Subject DiagnosisGFR

(mi/mm)Number ofnephrons

MeanSNGFR(nI/mm)

1 Normal 120 2 million 60

2 Patient A(chronic GN)

2 100,000 20

3 Patient B(interstitialnephritis)

2 33,400 60

a SNGFR refers to single-nephron glomerular filtration rate.

Table 2. Details of adaptation in sodium excretion: Nephronsubgroups

Subject DiagnosisNephronsubgroups

Mean valueSNGFR/nephron

(nl/min)

1 NormalABC

506070

2Patient A

(chronic ON)ABC

102030

3Patient B

(interstitialnephritis)

ABC

306090

rates per nephron in the normal subject differ to an astonishingdegree from the values in both patients. Remarkable differencesalso are evident between patient A and patient B. Finally, in thenormal subject, the values for each subgroup are somewhatdifferent from one another, and in the 2 patients the variationamong subgroups is striking. Yet in all 3 subjects, the single-nephron sodium excretion rates occur in response to theingestion of the same amount of salt, and, at the end of each 24-hour period, each subject excretes the same total amount ofsodium into the urine.

Variation in proximal tubular sodium reabsorptive rates. Thepotential problems confronting a biologic control systemcharged with maintaining sodium homeostasis become evenmore complex when, the calculated values for net sodiumreabsorption in different nephron segments of the individualsubgroups of each of the subjects are examined. Values forproximal reabsorption are shown in Table 3. We can assumethat homogeneity of glomerulotubular (G-T) balance is main-tained in all 3 subjects [13J and that by the end of the accessibleportion of the proximal tubule, all nephrons, irrespective oftheir subgroup, have reabsorbed 50% of the filtered sodium.Despite the equal values for fractional reabsorption, however,there are 7 different values for absolute sodium reabsorption forthe 9 nephron subgroups. These values range from 0.7 to 6.3nEq/min/nephron. Underscoring the seemingly random qualityof the mechanism we are attempting to analyze, the subgroup ofnephrons with the highest rate of sodium excretion per nephron

Table 3. Details of adaptation in sodium excretion: Proximal tubularsodium reabsorptiona

SubjectNephronsubgroup

SNGFR(ni/mm)

Proximal Sodiumreabsorption excretion

(nEqlmin/nephron)

NormalABC

506070

3.54.24.9

.035

.042

.049

Patient A(chronic ON)

ABC

102030

0.71.42.1

0.430.861.29

Patient B(interstitialnephritis)

ABC

306090

2.14.26.3

1.252.493.74

a 50% of the filtered sodium is assumed to be reabsorbed proximallyin all groups.

(patient B, subgroup C) simultaneously has the greatest rate ofproximal reabsorption of any of the 9 groups.

Variation in distal sodium reabsorptive rates. Values forsingle-nephron sodium reabsorption for the segment of thenephron beyond the accessible portion of the pars convoluta ofthe proximal tubule are shown in Figure 3. The subjects andsubgroups are the same as those in Figure 2. For each sub-group, the per-nephron rate of distal reabsorption is comparedwith the calculated value for sodium excretion. In all 3 subjects,tubulotubular (T-T) balance, as well as G-T balance, is assumedto remain homogeneous throughout the entire nephron popula-tion. Based on the calculated rates of distal sodium delivery(50% of the filtered load) and sodium excretion, 90% of thedistal sodium load is reabsorbed by the nephrons of the normalsubject, 38% by the nephrons of patient A (with chronicglomerulonephritis), and 40% by the nephrons of patient B(with interstitial nephritis). If lower values for fractional reab-sorption in the proximal tubule are used in one or both patients,the values for absolute distal reabsorption will increase, but thedispersion of values among the different subgroups will bequalitatively the same.

In the normal subject, the degree of variation in reabsorptionrates among the three subgroups is moderate. Within the 6subgroups of the 2 patients, however, distal reabsorption ratesvary over a range of almost tenfold. Moreover: (1) the meanvalue is different in each of the 6 subgroups; (2) the overallmean value (i.e., for the 3 subgroups combined) in patient B is 3times the comparable value in patient A; and (3) in all 6subgroups, the values for distal reabsorption are less than thelowest value in any of the subgroups in the normal subject. I willcome back to the last observation later when considering thepossible role of aldosterone in the control of sodium excretionin chronic renal disease.

The effect of a change in sodium intake. The physiologicresponse to a change in salt intake adds further to our difficultyin interpreting the characteristics of the control system. Table 4depicts the effects of a modest increase in sodium intake (from60 to 120 mEq/day) on single-nephron sodium excretion rates.To simplify the presentation, data are shown for only 2 nephronsubgroups, group A in the normal subject and group C in patientB. In response to an identical change in salt intake, equal to


4.4.

3.6

3.2

2.8

2.4

SNG FR

Fig. 3. Distal sodium reabsorption compared with sodium excretion pernephron (nEqimin). Six bars are shown for each of the 3 subjects. Foreach set of two bars, the one on the right represents sodium excretionper minute per nephron and the values are the same as those shown inFigure 2. The left-hand bars represent the calculated values for netsodium reabsorption beyond the last point of the "proximal tubule"accessible to micropuncture. The assumption is made that 50% of thefiltered sodium was reabsorbed proximally in each of the three subsetsof all 3 subjects. The effects of using different values for fractionalreabsorption in the 2 uremic patients are commented on in the text.

3.5 g of sodium chloride per day, the increase in sodiumexcretion per nephron is 100 times greater in the patient'snephron than in that of the normal person. It might be notedthat if all 2 million nephrons in the normal subject responded asdid the group C nephrons in patient B, the 24-hour sodiumexcretion would greatly exceed the total amount of sodium inthe extracellular fluid. Obviously, life would terminate longbefore the 24 hours passed. On the other hand, life persists inthe patient because of the magnified natriuretic response.

Let me reemphasize that: (1) A wide range of values exists forproximal sodium reabsorption; (2) there is a wide range ofvalues for distal sodium reabsorption; and (3) an especially widerange of values is found for overall sodium excretion pernephron. These differences obtain when one compares onepatient with the other, and they obtain when one compares anindividual nephron subgroup with another. Yet all 3 subjectsexcrete identical amounts of sodium in their urine daily. Thesystem that subserves the maintenance of sodium balancewould seem to possess remarkable, if not mysterious, qualities.

AdaptationFor sodium homeostasis to be maintained in advancing

chronic renal disease, the addition of any given amount ofsodium to the extracellular fluid must progressively raise the

Sodiumexcretion(nEqim in!nephron)

Total

SubjectGFR

(mI/mm) NephronsSNGFR(ni/mm)

60 mEq 120 mEqdieta dieta

Normal' 120 2,000,000 50 .018 .035 .018

Patient BC 2 33,400 90 1.87 3.74 1.87(interstitialnephritis)

average rate of sodium excretion per nephron as GFR falls. Wehave termed this central feature of the adaptation in sodiumexcretion in chronic renal disease the "magnification phenome-non" [14].

Of the many examples of the magnification phenomenon,perhaps the most graphic is the continued ability of patientswith chronic renal disease to maintain external sodium balanceon an unrestricted salt intake as GFR falls from normal toextremely low levels [15]. To accomplish this on a relativelyconstant sodium intake, FENa% must double with each 50% fallin GFR [16]. In Figure 4, values for whole-kidney FENa% areshown for the same 3 subjects described earlier (the FENa%values represent 24-hour averages). On the identical sodiumintake of 120 mEq/day, the normal person excretes 0.5%of thefiltered sodium (i.e., 1 of 200 filtered sodium ions), whereas ineach of the 2 patients, FENa% equals 32%. Presumably, thesevalues for FENa% are the same for all the nephrons in theindividual subjects regardless of SNGFR values.

Another example of the magnification phenomenon is seenwhen extracellular fluid is translocated from one portion of thebody to another without changing total extracellular fluidvolume. If a human is immersed in a tank of water to the level ofthe neck, there is a shift of extracellular fluid from the legs intothe upper portions of the body, In normal individuals, theincrease in central and intrathoracic extracellular fluid volumeis attended by a natriuresis [17]. When patients with chronicrenal disease of varying severity are subjected to water immer-sion, not only does sodium excretion increase, but the lower thesteady-state GFR, the greater is the rise in FENa% (Epstein M,Hoffman D, DeNunvio AG, unpublished observations). Thesepatients are nonedematous, in external sodium balance, andpresumably the volume of fluid translocated is closely compara-ble irrespective of the level of the GFR.

The phenomenon of "de-adaptation"The typical patient with chronic renal disease exhibits a salt-

losing state when GFR falls below 30 ml/min. In most instances,this condition is characterized by the patient's inability toreduce the obligatory sodium excretion in the urine to less than

NormalGFR 120 mI/mm

2,000,000 nephrons

Chronic G NGFR 2 mI/mm

100,000 nephrons

ISNGFR 2 mI/mm

33,400 nephrons

Table 4. Details of adaptation in sodium excretion:increasing sodium intake from 60 to 120 mEq/day on

sodium excretion

The effects ofsingle-nephron

The values for sodium intake are for 24-hour periods.b The data for the normal person are from Subgroup A (Table 2 and

Figure 2).The data for Patient B are from Subgroup C (Table 2 and Figure 2).


32

28

24

20

16

12

8

4

Normal Chronic GN ISNGFR 120 mI/mm GFR 2 mI/mm GFR 2 mI/mm

2,000,000 nephrons 100,000 nephrons 33,400 nephrons

Fig. 4. Fractional excretion of sodium when the intake is 120 mEqiday.The value for fractional reabsorption of sodium in each subject repre-sents the composite mean for all functioning nephrons (i.e., the valuesfor the three subsets have been averaged). As discussed in the text, thepreservation of homogeneity of G-T and T-T balances would make thevalues for fractional (but not absolute) excretion the same in all threenephron subsets of each subject. The calculated values are for a 24-hourperiod and assume that the total amount of sodium excreted is equal tothe total amount ingested (i.e., 120 mEq/day).

approximately 30 mEq/24 hours despite restriction of the di-etary sodium intake to less than this level [18]. Rarely, theobligatory sodium loss can be much higher, exceeding 200 to300 mEq/day [19, 20]. If the sodium intake is diminished belowthe "floor" of sodium excretion, negative sodium balance willensue, and a contraction of extracellular fluid volume necessari-ly will follow,

The general explanation for the salt-losing state of chronicrenal disease has varied from the dictates of electrical neutralityimposed by the excretion of unreabsorbable anions to a tubulardefect in sodium reabsorption. It also has been thought thatthere is a limitation in the lower limit of sodium concentrationachievable in the urine and thus that the magnitude of sodiumloss varies with urine flow [18]. An entirely different explana-tion for the salt-losing has emerged from recent studies byDanovitch, Bourgoignie, and Bricker [20]. Patients with chronicrenal disease whose GFRs ranged from 5.2 to 16 mI/mm weremaintained in a metabolic balance ward from 4 to 14 weeks. Theinitial sodium intake, based on a detailed dietary history, variedfrom 58 to 342 mEq/day. In each subject, salt intake wasreduced by small amounts at intervals of no less than one week.At the end of the study period, all the patients could maintainsodium balance on a diet containing 5.0 2.9 (SD) mEq/day.Moreover, there was no reduction in GFR or in estimatedplasma volume, nor were any adverse effects noted. The salt-losing state in these patients thus was reversed by slow andgradual "weaning" of sodium intake. The explanation proposedwas that the adaptation in sodium excretion in chronic renaldisease that results in the high rates of sodium excretion pernephron at low GFRs is not rapidly suppressible, but that "de-adaptation" can be induced if sodium withdrawal is accom-plished slowly over a long period. The patient under discussiontoday is a good example of this de-adaptive response.

Certain of the effects of de-adaptation on the magnificationphenomenon recently have been studied by Rever, Licht, and

Bricker (unpublished observations). Using the same salt-wean-ing technique just described, we subjected patients with chronicrenal disease to water immersion before and after the sodium-losing state was reversed. The resulting data clearly demon-strate that the magnified natriuretic response to water immer-sion in adapted patients with chronic renal disease is bluntedfollowing the de-adaptation process (Fig. 1). Measurements ofplasma volume before and after de-adaptation showed nochange. Our present interpretation is that the magnificationphenomenon for sodium depends on the preexistence of anadaptive natriuresis per nephron, but further documentation ofthis thesis is necessary.

General design of the sodium control system

I already have emphasized the view that a sophisticatedbiologic control system seems to be required for the mainte-nance of sodium balance. I also believe that although qualitativechanges in the mode of operation of this system occur continu-ously as GFR falls, the basic components of the system must bethe same in a patient with advancing chronic renal disease as ina normal person. If these views are correct, it follows that eachperson not only is born with a sodium control system, but thatthe system is designed so as to alter its patterns of functionprogressively in any individual who becomes afflicted with anephron-destroying process.

The control system is believed to contain a minimum of threecomponent parts [15]. The first is a detector element capable ofsensing an alteration in body fluids produced by the addition orloss of sodium chloride. The second is a transmittingelementcapable of relaying information about the perturbation of theextracellular fluid to the nephrons and altering their rate ofsodium transport. The third is the endorgan, which consists ofthe total pool of functioning nephrons whether the number be 2million or 30,000. There also is a possibility that an integratorelement may exist if there is not a single detector element, butrather widely dispersed detector elements. The integrator,presumably located in the central nervous system, wouldreceive the input from all detector elements, collate the data,and determine whether the message sent to the kidneys via thetransmitter is to increase or to decrease the rate of sodiumexcretion. An integrator would be of special value in thepresence of internal translocations of volume such as waterimmersion wherein there is a decrease in the effective extracel-lular fluid volume in the legs and a simultaneous increase in thethorax.

The detector element. Neither the location nor the nature ofthe detector element is yet known. It seems most likely,however, that the perturbation that is monitored is a change inextracellular fluid volume or a hemodynamic alteration inducedby a volume change, rather than a change in the extracellularfluid sodium concentration per Se. Thus, the sodium controlsystem in fact is probably a volume control system, and sodiumbalance is maintained by virtue of the usually close couplingbetween changes in extracellular fluid sodium content andextracellular fluid volume.

Several lines of evidence suggest that the detector element islocated in the upper half of the body, possibly within the thorax,perhaps in the brain. In addition to the water immersion studies,space flight is associated with an early natriuretic response; thechange from the l-g environment of earth to the 0-g environ-


ment of space is attended by the movement of approximately 2liters of fluid from the lower extremities into the more cephaladportions of the body [211. Prolonged bedrest also leads to acephalad shift of extracellular fluid and typically produces anatriuresis. Movement of extracellular fluid volume from theupper to the lower portions of the body, such as is induced byquiet standing and application of thigh tourniquets, alsochanges sodium excretion, but in the opposite direction [221.

The integrator element. The basis for considering the exis-tence of a centrally located "computer"-type mechanism thatwould monitor simultaneous, and in some instances opposing,changes in regional extracellular fluid volumes already has beenalluded to. If such an integrator element exists, however, itslocation and mechanism of action remain unknown.

The transmitting element. There is an extensive and growinglist of factors known to influence sodium excretion. Amongthese are: (1) changes in GFR throughout the nephron popula-tion; (2) changes and/or redistribution of GFR values in corticalversus juxtamedullary glomeruli; (3) changes in medullaryblood flow with an alteration of sodium concentration gradientsbetween the medullary interstitium and the vasa recta and loopof Henle; (4) changes in sympathetic nervous activity; and (5)alterations in the activity of one or more of a number of humoralsubstances including aldosterone, prostaglandins, vasopressin,oxytocin, the kallikrein-kinin system, parathyroid hormone,calcitonin, and finally the putative natriuretic hormone.

For any one of the foregoing factors to be established as thekey modulator of sodium excretion, its activity must be shownto correlate closely and consistently with both short- and long-term changes in sodium excretion under many different circum-stances, including chronic renal disease. None of the factorslisted has yet been examined under enough circumstances toestablish which, if any, is the principal modulator of sodiumexcretion. It does appear, however, that most of the factorslisted can be excluded on the basis of negative correlations. Theevidence in regard to aldosterone and natriuretic hormone willbe considered later.

The end organ: The nephron. A vast amount of data relatingto the manner in which sodium is handled by the nephron hasbeen accumulated, and most of this information falls beyond thescope of this discussion. But I would like to comment a littlelater about the possibility that intrinsic changes in nephronfunction might contribute to the magnification phenomenon.

Special comments about aldosterone and natriuretic hormoneAldosterone. Clear and incontrovertible evidence shows that

aldosterone augments epithelial cell sodium transport capacity.However, the biologic role of aldosterone in regulating tubularreabsorption of sodium seems to involve the induction of thesynthesis of proteins concerned with sodium transport ratherthan the modulation of the minute-to-minute rate of sodiumtransport and excretion [23]. In chronic renal disease, thepattern of change in aldosterone levels is not consistent as GFRfalls from normal to low levels. Elevated levels have beenreported, yet in the data shown in Figure 3, values for distaltubular reabsorption in the 2 patients were lower in all nephronsubgroups than in any of the 3 subgroups in the normal subject.The values were particularly low in the patient with glomerulo-nephritis. A number of experimental observations also fail tosupport the view that aldosterone is the primary "transmitter"

in the sodium control system in chronic renal disease [16, 24,25].

Natriuretic hormone. Although natriuretic hormone has notyet been isolated and synthesized, the body of evidence sup-porting its existence is compelling. Much of this evidence hasbeen reviewed in a recent monograph [26]. I will limit mycomments about natriuretic hormone to a brief summary ofcertain of the more relevant data. The natriuretic factor, theterm I generally prefer to use rather than natriuretic hormone, isa low-molecular-weight compound (approximately 500 daltons),which has been separated from the serum and urine of uremicpatients who have high fractional sodium excretion rates [27,28]. Because natriuretic factor is present in increased amountsin both the serum and urine of uremic patients, one can assumethat its rate of production is increased in uremia. Natriureticfactor is not detectable, however, in the urine or serum ofpatients with advanced chronic renal disease who are nephroticand edematous and, hence, are not undergoing a natriuresis[28]. A sodium transport inhibitor, probably the same substancepresent in uremia, also has been found in: (I) normal individualsingesting a high-salt diet [29]; (2) healthy dogs ingesting 15 g ofsodium chloride and 0.2 mg of fludrocortisol per day followingescape [30]; (3) normal subjects during water immersion [31];(4) patients with aldosterone-secreting tumors (sodium trans-port inhibition disappears following removal of the tumor) [32];and (5) patients with head trauma who are manifesting anatriuresis [33].

As in nephrotic uremic patients, natriuretic activity was notdemonstrable with standard bioassay techniques in uremic dogsin which the typical natriuresis per nephron was prevented by areduction in dietary sodium intake in exact proportion to thereduction in GFR [34]. Activity also was absent in patients withhead trauma in whom a natriuretic state did not develop [33].We have studied 2 patients with chronic renal disease in whomassays for natriuretic activity in the urine were inexplicablynegative; but in both the assays became positive during waterimmersion (Rever B, Licht A, Bricker NS, unpublishedobservations).

Natriuretic factor inhibits sodium transport by the isolatedtoad bladder [28], the isolated frog skin [28], the isolatedperfused cortical collecting tubule of the rabbit [35], and twocell types grown in tissue culture (Licht A, Bricker NS,unpublished observations). One of these cell lines, the so-called"MDCK" strain, originally was derived from tubular epithelialcells of the dog. The other, the 3T6 strain, is a line of fibroblaststhat actively transports sodium. Finally, natriuretic factor pro-duces a sodium diuresis both in the normal water-loaded rat [36]and in the uremic rat fed a high-salt diet for 36 to 48 hoursbefore the material is administered [28].

In the isolated tubule preparation, toad bladder, and the frogskin, natriuretic factor is active only when added from thebasolateral, or the "blood," surface [37]. It increases theintracellular sodium content of isolated toad bladder epithelialcells and decreases their rate of pyruvate oxidation [38].Natriuretic factor inhibits sodium efflux in the isolated perfusedtubule but does not affect sodium influx [35]. It inhibits sodiumefflux by both the MDCK and the 3T6 cell systems and does notproduce further inhibition of sodium efflux in the same celltypes incubated in l0 M ouabain (Licht A, Bricker NS,unpublished observations). It inhibits rubidium influx in the 3T6


system (Licht A, Bricker NS, unpublished observations), but itis not kaliuretic in the rat [27, 281.

Intensive efforts are in progress to isolate natriuretic factor,and a relatively high degree of purity has been obtained usinghigh-performance liquid chromatography. The active fraction iswater soluble, soluble in some organic solvents [36, 39] but notin others, and presumably is polar in nature. Natriuretic factorappears in a chromatographic peak (using high-performanceliquid chromatography) that contains an active amine group; itis inactivated by some proteolytic enzymes but not by others,and disagreement still exists as to whether it is a peptide. Manyof its effects resemble those of ouabain, and it has beensuggested that natriuretic factor might be an endogenous digital-is glycoside-like compound [40].

Operation of the control system: A hypothetical modelThe model to be presented fits within the classic framework

of a "detector-transmitter-end organ" system that maintainssodium homeostasis by controlling the constancy of effectiveextracellular fluid volume. Given the changing circumstancesinherent in advancing renal insufficiency, how does such asystem continue to operate? In essence, how can a specificamount of sodium, an amount that can vary from meal to mealand day to day, be excreted with some precision by a popula-tion of nephrons that varies in number from 2 million to perhaps30,000 and that can exhibit the spectrum of values for proximaland distal sodium reabsorption and sodium excretion per neph-ron, exemplified by Figures 2 and 3 and Table 3? We can focusthis question around two related issues: (1) is there a mecha-nism that serves as the "control center," overseeing thecontinuous operation of the system and overriding seeminglyinsurmountable obstacles; and (2) how can the magnificationphenomenon be explained?

The "control center." Theoretically, the system could be"effector oriented," that is, have its control center located inthe kidneys per se, or it could be "detector oriented," and havethe control center located anywhere in the body. The firstpossibility would require that the kidneys possess a means ofdetermining how much sodium is acquired daily and thus howmuch must be excreted. Following the addition of sodium to theextracellular fluid, excretion per nephron then would have to beincreased, presumably by a reduction in distal sodium reabsorp-tion [35]. The reabsorption of sodium might emit some bio-chemical signal and a feedback loop then might modulate therate of release of a transmitter element and evoke a subsequentdecrease (or increase) in distal sodium reabsorption. However,a model that places the control center in the kidneys leaves amajor problem unsolved. For the end organ to monitor the rateof addition (or deletion) of sodium to the extracellular fluid, asensing device would be required. Despite the intrarenal loca-tion of such a hypothetical sensor, it nevertheless would bemonitoring a volume-related, presumably hemodynamic,change. In this sense, such a control system would be just as"detector oriented" as it would be if it resided in the chestrather than in the renal parenchyma. Thus, we believe thevolume control system can be viewed as detector oriented withits "control center" probably residing in the detector elementper se. Let us examine a working model of the system havingthe general design presented in Figure 5.

The model control system is designed to maintain an optimal

Basal state

ECF_))uTseDetector

NaCI ingestion

Postprandialovershoot /

extracellular fluid volume. The upper portion of the figuredepicts the basal state, wherein the volume of the extracellularfluid is normal. The detector element receives little input andsends out only minimal pulses to the site of synthesis of themajor transmitter element, which we will assume for thepresent discussion is natriuretic hormone. In health, in the basalstate, the level of the hormone in the circulation is low, thedegree of inhibition of distal tubular sodium reabsorption issmall, and urinary sodium excretion is minimal. This basal statewould prevail some hours after the ingestion of a meal andperhaps would be most characteristic of the pattern observedduring early morning.

The middle panel in Figure 5 depicts the effects on the controlsystem of the ingestion of sodium chloride. The entrance of theNaC1 into the extracellular fluid leads to isosmotic expansion ofthe extracellular fluid as previously noted. The detector ele-ment then senses the expansion through a still-to-be-definedhemodynamic event induced by the expansion, and the activityof the detector is markedly increased over that which prevailedin the basal state. The result is that the pulses sent out to theorgan producing the transmitter are increased; the thick blackarrow in the figure depicts a marked increase in the rate ofrelease of the transmitter substance. The transmitter, whichagain is assumed to act by inhibiting distal tubular sodiumtransport, will, after reaching the nephrons via the renal circula-tion, diminish net sodium reabsorption per nephron and therebyincrease sodium excretion per nephron. The concentration ofthe inhibitor in the postglomerular blood should be identical forall nephrons (with the possible exception of the juxtamedullaryunits) and, if the number of receptor sites for the transmitter is

-

Fig. 5. Proposed model of the biologic control system for sodium. N. H.stands for putative natriuretic hormone.


proportional to SNGFR, and the affinity of the unoccupied sitesis constant from nephron to nephron, the number of newtransmitter molecules that will attach to the unoccupied recep-tors also should be proportional to SNGFR. Hence, the per-centage change in distal tubular sodium reabsorption of thedistal load of sodium would be identical in all nephrons andwould be determined by (1) the concentration (or activity) of thetransmitter element in the blood; (2) the affinity of the receptorsites for the inhibitor; and (3) the number of new receptor sitesoccupied. Despite the homogeneous change in the percentageof distal sodium reabsorption, as discussed previously, theactual number of sodium ions prevented from crossing theepithelial cells, and thus remaining in the tubular fluid to beexcreted, will vary from one nephron to another in the samekidney depending on the distal load of sodium, and the latterwill depend, in turn, on the value for SNGFR. The dispersion ofvalues for sodium excretion rate per nephron, as shown inFigure 2, thus will vary widely as does SNGFR. Nevertheless,the effect of the increased activity of the transmitter on theresidual nephrons, regardless of their number, will be to in-crease total sodium excretion.

The bottom panel represents the system's status after thetransmitter has exercised its effects on the nephrons. All theingested sodium chloride has been excreted, but the possibilityexists that an "overshoot" phenomenon might be involved. Itseems likely that as the extracellular fluid volume is restoredtoward normal, the stimulus to the detector element willdiminish pan passu. It also seems likely, however, that thesystem cannot operate with absolute precision and that theextracellular fluid volume will fluctuate about an "ideal" vol-ume. Thus, despite the fact that the full amount of addedsodium chloride has been excreted, the activity of the transmit-ter may persist at a level greater than basal for a finite periodand result in a continuing, albeit diminished, natriuresis. Theovershoot phase should be self-limited, and when additionalsodium chloride is ingested, the extracellular fluid volume willfirst return to the basal level and then undergo expansion.

In essence, therefore, with a detector-oriented model, thenatriuresis attendant on the addition of sodium chloride to theextracellular fluid will begin to abate as the extracellular fluidvolume begins to diminish toward normal. Activation of thedetector in this model plays a key role in determining the rate ofrelease of the transmitter element. The model provides for thepreservation of sodium homeostasis through the maintenance ofexternal balance regardless of the number of nephrons or of thedispersion of single-nephron sodium excretion rates.

The magnification phenomenon. The question must be askedwhether, in the progression from 2 million to 30,000 nephrons,the identical volume pertubation produces a progressivelygreater degree of activation of the detector element, therebyexplaining the magnification phenomenon. Although this possi-bility cannot be ruled out, it seems to me extremely unlikelythat each time a nephron is lost, a change will occur in thesensitivity of the detector element; for it should be noted thatthe magnification phenomenon is characteristic not only ofsodium, but also of phosphate, magnesium, potassium, andprobably a number of other solutes. The detector element forphosphate probably monitors a change in ionized calcium; thatfor potassium may monitor a change in transmembrane chemi-cal or electrochemical potential gradient; and that for magne-

sium perhaps monitors a change in the concentration of ionizedmagnesium, and so forth. If the magnification phenomenon foreven these few solutes were based on a stepwise and simulta-neous increase in sensitivity of each element, one would have toinvoke multiple separate mechanisms (one for each system)whereby each wave of nephron loss would "reset" the sensitiv-ity of each detector mechanism.

The evidence for an increase in production of natriureticfactor in chronic renal disease already has been cited. Whatmediates this increase is unknown, but there is no consistent orcompelling evidence to suggest that the change is due to acontinuous expansion of extracellular fluid volume throughoutthe course of chronic renal disease. There are no data, howev-er, concerning the rate of release of natriuretic factor inresponse to the addition of a given amount of sodium at eachlevel of GFR throughout the entire course of chronic renaldisease. Although such data will be quite difficult to obtain untila highly sensitive assay system is developed, they ultimatelywill be essential to have if the missing pieces of this puzzle areto be filled in. The possibility that the release rate of thetransmitter element varies as an inverse function of GFR inresponse to the ingestion of a fixed amount of sodium thuscannot be dismissed until the requisite data are obtained.Another possible explanation for the magnification phenome-non is that there is an increase in end-organ sensitivity to thetransmitter element and that this heightened response might bedue either to an enhanced intrinsic sensitivity or responsivity ofthe nephron and/or to an increase in the number of receptorsper unit of SNGFR or of tubular length. Indirect evidencesupporting increased end-organ sensitivity was obtained inexperiments in which a fixed amount of natriuretic factor wasinjected into the renal artery of normal and uremic rats. Theestimated rate of sodium excretion per nephron increasedsubstantially more in the uremic than in the normal animals[41].

In summary, the magnification phenomenon is essential forthe preservation of sodium homeostasis in chronic renal diseaseand thus for the preservation of life in the presence of progres-sive nephron loss. Evidence suggests that the activity of thetransmitter element, which inhibits sodium reabsorption by thedistal tubule, may increase in response to a given load ofsodium as GFR falls. Further, end-organ sensitivity mightincrease as chronic renal disease advances. This remarkablemagnification phenomenon, the central element of the adapta-tion in sodium excretion in chronic renal disease, remainsbasically unexplained, although many aspects of the phenome-non should be readily amenable to experimental examinationwith modern techniques.

Conclusions

We have examined the processes involved in the mainte-nance of homeostasis of a single solute, sodium, in healthyindividuals and in patients with chronic renal disease such asthe one presented today. But homeostasis also is maintainedwith close precision for a host of other solutes, and for some ofthese, the homeostatic process adapts continuously as diseasedestroys greater than 90% of the original nephron population. Ifa biologic control system does oversee sodium homeostasis, webelieve that biologic control systems probably exist for othersolutes as well. At least some of these control systems, and


perhaps all, could contain transmitter elements that modulatetubular transport of their specific solute. Some systems, such asthe one for sodium, however, would require a transmitter thatinhibits reabsorption; others would need a transmitter thatstimulates secretion; and it is conceivable that still others mightrequire a transmitter element that alters the balance betweenreabsorption and secretion. All the systems must be capable ofoverriding any interference produced by the action of othercontrol systems. In general, the rates of acquisition of keysolutes of body fluids vary both randomly and independently ofone another in advancing renal disease as well as in the healthystate. It is obvious that an increase in sodium intake cannot leadto a persistent increase in phosphate or potassium excretiondespite the fact that natriuretic factor has been shown to bephosphaturic and that natriuresis ordinarily increases potassi-um excretion; moreover, a decrease in sodium intake cannotlead to a reduction in phosphate excretion, potassium excre-tion, or the excretion of any other solute that is influenced bythe reabsorption of sodium by tubular epithelial cells.

The unraveling of the nature of solute-specific biologic sys-tems and the changes that take place in those systems in chronicrenal disease, I believe, represents one of the most interestingand important biomedical challenges facing us in the comingdecade. Indeed, I would submit that this area stands among thekey unsolved mysteries of mammalian biology.

Questions and answers

DR. BARTON LEVINE (Assistant Professor of Medicine,Wadsworth VA Medical Center, UCLA School of Medicine,Los Angeles, California): In what tubular segment or segmentsis natriuretic hormone thought to have an effect?

DR. BRICKER: There are recollection micropuncture data thatsuggest a modest effect on the proximal tubule [42], but it wouldappear that the principal effect is on the cortical collectingtubule [35].

DR. LEON G. FINE (Director, Nephrology Division, Centerfor the Health Sciences, UCLA School of Medicine, LosAngeles): The data you showed regarding the effects of waterimmersion are interesting. However, before one can concludethat these observations reflect a change in the sensitivity of theend organ, one must know that the movement of fluid is thesame with the patient on a low or high salt intake. You indicatedthat the measured blood volume was the same whether thepatient was on a high- or low-salt diet. I would predict that ifany part of the extracellular fluid composite was protected, itwould be the blood volume. There is a good chance, however,that the total extracellular fluid volume or the interstitialvolume would be significantly lower when the patient ingested alow-salt diet, and the differences you report could be explainedby differences in the amount of fluid shifted. Unfortunately, inwater immersion, neither the hematocrit nor a change in proteinconcentration can be used to reflect a change in extracellularvolume.

DR. BRICKER: First, I did not intend to imply that the site ofthe increased responsivity in the adapted state is necessarilylimited to the nephron. It could reside at any level, or multiplelevels, of the control system. With respect to whether "de-adaptation" was associated with extracellular fluid volumecontraction, I agree that the equality of plasma volumes in theadapted and de-adapted states is not in itself sufficiently strongevidence to rule out this possibility and that some index of

central blood volume, cardiac index, or plethysmographic mea-surements of the legs would be helpful. Extracorporeal volumeexpansion in the adapted versus the unadapted state also wouldhelp to clarify this question, and such studies are being conduct-ed by Dr. Michael Shapiro in our laboratories.

DR. GABRIEL M. DANOVITCH (Director, Clinical Nephrolo-gy, Center for the Health Sciences, UCLA School of Medicine,Los Angeles): We recently reported a patient who manifestedspontaneous changes in sodium excretion very similar to thoseseen in the two water immersion studies performed in thepatient under discussion [43]. Our patient suffered the unusualcombination of advanced polycystic renal disease and Crohn'sdisease. He had an ileostomy, from which he lost an apprecia-ble amount of sodium-containing fluid daily. His diet contained150 mEq of sodium per day, but he excreted only 5 or 6 mEq ofsodium in his urine; the rest was lost in the ileal fluid. Thisunique set of circumstances had prevailed for approximately 10years. Thus, as his renal failure advanced, he did not "need" todevelop an adaptive natriuresis per nephron and, in fact, heremained "unadapted," excreting virtually no sodium in hisurine. When he was subjected to marked extracellular fluidvolume expansion, his urine contained little or no sodium. Thushe did not exhibit the magnification phenomenon. We interpret-ed this as an experiment of nature in which "de-adaptation"occurred spontaneously. We believe the data are consistentwith the results of the water immersion studies, in which amagnified natriuresis failed to occur in the absence of adapta-tion. Later in his course, however, after prolonged volumerepletion, his sodium excretion increased.

DR. BRICKER: This fascinating experiment of nature bearsdirectly on the key question of whether a preexisting adaptationis necessary for the magnification phenomenon to work forsodium. In the case of magnesium (Kirschenbaum MA, Lie-bross B, Bricker NS, unpublished observations) and probablypotassium [44], adaptation does appear to be necessary formagnification to occur. In the case of phosphate, however, itdoes not [451.

DR. FINE: One could conceive of a hormone-mediated natri-uresis occurring without an increase in hormone concentration.If the number of nephrons decreased but an excess of hormonereceptors still existed, more hormone-receptor complexeswould be formed on those remaining nephrons, and a corre-spondingly greater natriuresis per nephron would ensue. Butwhen you measure natriuretic hormone with existing bioassaytechniques, the hormone concentration in serum is elevated;the question is why. Could this increase be related to "steady-state" blood volume? How adequate are the data on measuredblood volume in uremic patients? If the hormone concentrationis elevated simply because uremic patients have higher bloodvolumes than do nonuremic patients, then that elevation couldsimply reflect a different steady state and it might not play a rolein governing the excretion of sodium.

DR. BRICKER: Your hormone-receptor theory, if validated,could help explain the magnification phenomenon. One mightexamine whether the isolated perfused cortical collecting tubulefrom normal versus uremic, and adapted uremic versus un-adapted uremic, animals exhibits differences in sensitivity tothe same amounts of natriuretic factor in vitro.

The question of whether blood volume or, more importantly,extracellular fluid volume is consistently elevated in uremia isbeclouded somewhat by methodologic problems. Thus the


existing methods, particularly those used for measuring extra-cellular fluid volume, are not sufficiently precise to allow for adefinitive answer. Some years ago, Schultze studied normaldogs in which he measured inulin space and blood volume. Hethen made the dogs uremic, maintained the same salt intake,and repeated the space measurements; he found no changes[24]. More recently Dr. Michael Shapiro has found that inulinspaces are no greater in uremic than in normal rats and nogreater in sodium-adapted than in sodium-unadapted uremicrats (unpublished observations). However, the conclusionsdrawn from these observations must be tentative because theerror in inulin space measurements must be at least 5%.

DR. JOEL D. KOPPLE (Professor of Medicine and PublicHealth, Wadsworth VA Medical Center, UCLA School ofMedicine, Los Angeles): A shift of 700 ml of fluid into the upperhalf of the body during water immersion represents less than 3%of total body water. This is well below the accuracy of thesemeasurements of extracellular space.

I have a question, Dr. Bricker. I find the welter of datastrongly suggestive that there is a humoral factor. But I findproblematic the length of time it takes to de-adapt a uremicpatient from the presumed effects of this hormone. With everyother hormonal system I am aware of, the adaptive response ismeasured in terms of minutes or, at the most, hours; yet in thisinstance a patient can become profoundly volume depleted,including having marked reductions in blood volume, and stillhave inappropriate sodium excretion. Granted, it may be be-cause not only is the circulating hormone increased but alsobecause the receptors are increased. But don't you find ittroublesome that it takes so many weeks to wean a patient fromthe natriuretic state?

DR. BRICKER: Yes. But there is at least one analogoussituation, namely, the uremic patient with poorly suppressiblehyperparathyroidism, in whom it may take a year or longerbefore the euparathyroid state is restored after successful renaltransplantation. Also the salt-losing state does not seem tooccur in uremic rats or dogs, possibly because the adaptation inlaboratory animals generally evolves over a very much shorterperiod of time than in patients.

DR. JORDAN J. COHEN: Have you ever failed in an attempt todc-adapt a patient? That is, do some uremic patients haveunremitting salt wastage?

DR. BRICKER: Dr. Rever, would you like to respond to that?DR. BARBARA L. REVER (Assistant Professor of Medicine,

Center for the Health Sciences, UCLA School of Medicine, LosAngeles): I have studied 5 patients and have encountered onlyone in whom we could not decrease the daily sodium intakebelow 15 mEq. Dr. Danovitch has studied 5 patients and wassuccessful in all [20]. Thus dc-adaptation has been accom-plished successfully in 9 of 10 patients.

DR. AMNON LICHT (Assistant Professor of Medicine, Centerfor the Health Sciences, UCLA School of Medicine, LosAngeles): When de-adaptation is attempted too quickly, volumedepletion can develop, requiring saline infusion. However, witha slower rate of sodium "weaning," the same patient can be de-adapted successfully. This was apparent in today's patient.

DR. KIY0sHI KUROKAWA (Chief, Nephrology Division,Wadsworth VA Medical Center, UCLA School of Medicine,Los Angeles): In uncomplicated patients with chronic renalfailure who ingest a stable sodium intake, have you found agood correlation between measured plasma volume or plasma

renin activity on the one hand, and the level of sodium intake onthe other?

DR. BRIcKER: Renin values decrease during water immersionin uremic patients. However, I am not aware of a systematic setof data that would permit a clear answer to your question.

DR. KUROKAWA: If this sensor system is to be operative inthe presence of chronic renal failure, say in a patient with aGFR of 10 ml/min and a sodium intake of 100 mEq/day, a smallincrease in sodium intake, of say 20 mEq/day, must result in anequivalent increase in sodium excretion, as would occur innormal individuals. Given that patients with reduced GFRs arethought to have high levels of natriuretic factor to start with, itis unclear how these patients can detect relatively small addi-tional increments in volume and translate these increases intothe appropriate change in sodium excretion. Because thesepatients have high levels of natriuretic factor to start with, it ishard for me to accept that the same small increase wouldproduce the same increment in sodium excretion as it does innormal individuals.

DR. BRICKER: In preliminary studies, again by Dr. MichaelShapiro, natriuretic factor has been found in the urine ofadapted uremic rats. It is highly likely that the animals used forthe bioassay had high endogenous levels of natriuretic factor.The addition of 20 mEq/day of sodium, barring different rates ofexcretion, would produce the same increment in extracellularfluid volume in the patient as in the normal person; thus theincrease in endogenous natriuretic factor should be equal.

DR. COHEN: What, if any, influence might an altered diurnalpattern of sodium input and output have on sodium adaptation?After all, we don't ingest sodium continuously but rather inboluses with meals throughout the day. It is conceivable thatthe response of a normal individual and of a patient with markedrenal insufficiency to the same increment in sodium intakecould be identical when expressed as mEq/24 hours but quitedifferent when looked at in temporal profile over a typical day.Are there any data about the relative rates at which a givensodium increment, provided in bolus form, is excreted by anormal person versus a patient with chronic renal failure?

DR. BRICKER: There are both animal and human data, butthere are contradictions. In one study, uremic dogs who re-ceived a 2.5 g sodium chloride load excreted approximately thesame amount of sodium in 5 hours as did the same dogs beforethey became uremic [24]. More recent studies from St. Louisshowed a slower short-term rate of excretion in uremic dogs[46]. Finally, in studies currently being performed, Danovitchand Licht have found a consistent increase in excretion ofnatriuretic factor in normal individuals between midnight and4:00 A.M.

DR. MICHAEL A. KIRSCHENBAUM (Assistant Professor ofMedicine, Center for the Health Sciences, UCLA School ofMedicine, Los Angeles): The elusiveness of the identification ofthe components of the volume regulation system is very dis-turbing. In many other biologic control systems, it has been agreat deal easier to identify components. Yet in primate as wellas subprimate studies seeking the elements of the detector limb,negative results have been obtained; these failures suggest thatperhaps the detector, controller, and effector are located intra-renally. Maybe we are looking at a purely intrarenal eventwhich, as Dr. Fine suggests, might overshoot substantially andwhich in most situations might not be very selective orsensitive.


DR. BRECKER: I can only reiterate the fact that the lower theGFR, the greater the change in tubular epithelial transport (atany given value for SNGFR) for a given change in intake of asingle solute, and that the rates and direction of change forsodium, as well as for other substances, are solute specific andgeared to the preservation of homeostasis. Also, the ability to"wean" a chronically uremic patient of dietary sodium with noconsistent change in plasma sodium concentration, plasmavolume, or GFR bespeaks a highly ordered control system.

DR. KUROKAWA: Dr. Fine has demonstrated that natriureticfactor modulates sodium transport in the cortical collectingtubule. It should be feasible, therefore, for one to compare theresponse of tubules obtained from animals with chronic renalfailure with th&se obtained from normal controls. If one pro-duces a dose-response curve using the same source of natriuret-Ic factor in serial dilutions, shouldn't one be able to demon-strate differences in the sensitivity—that is, a shift of theresponse curve to the right or left—as well as differences in theresponsiveness—that is, the magnitude—of tubular response?

DR. BRICKER: We are in the process of studying the natriuret-Ic effect of urine fractions from adapted versus de-adapted ratsusing adapted and de-adapted rats for bioassay. The studies yousuggest are important and I hope they will be done soon.

But let me make one other point with regard to your question.One can make dogs uremic by reducing nephron mass in serialsteps; if salt in the diet is decreased in exact proportion to thedecrease in GFR, such animals will become uremic and have noincrease in natriuresis per nephron; moreover, the bioassay fornatriuretic hormone will remain negative [34]. Your questioncould be answered by studying isolated perfused cortical col-lecting tubules from such nephrons and comparing the results tothose obtained in nephrons from uremic animals with highlevels of natriuretic factor.

DR. HECTOR J. RODRIGUEZ (Assistant Clinical Professor ofMedicine, UCLA School of Medicine, Los Angeles): You haveincluded aldosterone in your scheme as a modulator of sodiumexcretion. Does available evidence permit one to excludechanges in sensitivity or responsiveness of the surviving neph-rons to aldosterone levels?

DR. BRICKER: I think aldosterone plays a very important rolein sodium homeostasis. But experimental observations suggestthat if one maintains mineralocorticoid hormone levels in auremic dog or patient at a fixed level, whether this be very lowor very high, and if one modifies the sodium intake, sodiumbalance will be maintained [16, 24].

DR. FINE: Gruber, Whitaker, and Buckalew recently report-ed a circulating ouabain-like substance [40]. Do you think theyhave discovered the natriuretic hormone?

DR. BRICKER: Although samples have not been exchangedbetween laboratories—and in my view, there is an urgent needthat this be done—they believe that their inhibitor can bemeasured using a radioimmunoassay for digoxin. I don't knowwhether their substance is natriuretic hormone, but, as I notedearlier, natriuretic hormone could be an endogenous, digitalisglycoside-like compound. Many characteristics of the hormoneresemble those of ouabain: the hormone is active only whenadded to the basolateral or blood surface, it increases intracellu-lar sodium content, it blocks sodium efflux but doesn't affectsodium influx, it decreases oxidative phosphorylation, it has theright molecular weight, and there are other similarities.

DR. FINE: Let's say then that Gruber, Whitaker, and Bucka-lew do have a circulating natriuretic substance. Does one stopthere and say the natriuretic hormone has been discovered, ordoes one go on looking for another natriuretic hormone?

DR. BRICKER: I'd be happy with one!

AcknowledgmentsThe original work cited in this paper was supported by USPHS AM

25287, USPHS AM 26098, NASA NCC2-104, and the BurroughsWellcome Fund.

Reprint requests to Dr. Neal S. Bricker, Program in Kidney Dis-eases, UCLA School of Medicine, Los Angeles, California 90024, USA

References

1. BANK N, AYNEDJIAN HS: Individual nephron function in expen-mental bilateral pyelonephritis, I. Glomerular filtration rate andproximal tubular sodium, potassium, and water reabsorption. fLabClin Med 68:713—727, 1966

2. LuBowITz H, PURKERSON ML, SUGITA M, BRICKER NS: GFR pernephron and per kidney in the chronically diseased (pyelonephritic)kidney of the rat. Am J Physiol 217:853—857, 1969

3. ROCHA A, MARCONDES M, MALNIC G: Micropuncture study in ratswith experimental glomerulonephritis. Kidney mt 3:14—23, 1973

4. LUBOWITZ H, MAZUMDAR DC, KAWAMURA J, Ciosso JT,WEISSER F, ROLF D, BRICKER NS: Experimental glomerulonephri-tis in the rat: Structural and functional observations. Kidney mt5:356—364, 1974

5. KRAMP RA, MACDOWELL M, GOTTSCI-IALK CW, OLIVER JR: Astudy by microdissection and micropuncture of the structure andthe function of the kidneys and the nephrons of rats with chronicrenal damage. Kidney mt 5:147—176, 1974

6. SCHULTZE RG, WEI5SER F, BRICKER NS: The influence of uremiaon fractional sodium reabsorption by the proximal tubule of rats.Kidney liii 2:59—65, 1972

7. HAYSLETT JP, KASHGARIAN M, EPSTEIN FH: Mechanism of thechange in the excretion of sodium per nephron when renal mass isreduced. J Clin Invest 47:774—782, 1968

8. ALLISON MEM, WILSON CB, GOTTSCHALK CW: Pathophysiologyof experimental glomerulonephritis in rats. J Clin Invest 53:1402—1423, 1974

9. BLANTZ RC, WILSON CB: Acute effects of antiglomerular base-ment membrane antibody on the process of glomerular filtration inthe rat. J Clin Invest 58:899—911, 1976

10. BLANTZ RC, TUCKER BJ, WILSON CB: Acute effects of anti-glomerular basement membrane antibody on the process of glomer-ular filtration in the rat. Influence of dose and complement deple-tion. J Clin Invest 61:910—921, 1978

11. WEBER H, LIN K, BRICKER NS: Effect of sodium intake on singlenephron glomerular filtration rate and sodium reabsorption inexperimental uremia. Kidney mt 8:14—20, 1975

12. DEEN WM, MADDOX DA, ROBERTSON CR, BRENNER BM: Dynam-ics of glomerular ultrafiltration in the rat. VII. Response to reducedrenal mass. Am J Physiol 227:556—562, 1974

13. BRICKER NS, FINE LG: The renal response to progressive nephronloss, in The Kidney (2nd edition), edited by BRENNER BM, RECTORFC JR, Philadelphia, W. B. Saunders, 1981, pp. 1056—1096

14. BRICKER NS, FINE LG, KAPLAN MA, EPSTEIN M, BOURGOIGNIEJJ, LICHT A: "Magnification phenomenon" in chronic renal dis-ease. NEnglJMed299:1287—1293, 1978

15. BRICKER NS, DANOVITCH GM: Extracellular fluid volume regula-tion: On the evidence for a biologic control system, in The Kidneyin Liver Disease, edited by EPSTEIN M, New York, Elsevier/NorthHolland, in press

16. SLATOPOLSKY E, ELKAN 10, WEERT5 C, BRICKER NS: Studies onthe characteristics of the control system governing sodium excre-tion in uremic man. J Clin Invest 47:521—530, 1968

17. EPSTEIN M: Cardiovascular and renal effects of head-out waterimmersion in man. Circ Res 39:619—628, 1976


18. COLEMAN AJ, ARIAS M, CARTER NW, RECTOR FC, SELDIN DW:The mechanism of salt wastage in chronic renal disease. J C/inInvest 45:1116—1125, 1966

19. STANBURY SW, MAHLER RF: Salt wasting renal disease: Metabolicobservations on a patient with "salt-losing nephritis." Q J Med28:425—447, 1959

20. DANOVITCH GM, BOURG0IGNIE JJ, BRICKER NS: Reversibility ofthe "salt-losing" tendency of chronic renal failure. N Engl J Med296:14—19, 1977

21. BRICKER NS: Life Beyond the Earth's Environment: The biology ofliving organisms in space, in Monograph for Space Science Board,edited by BRICKER NS, Washington, D.C., National Academy ofSciences, 1979

22. FITZHUGH FW JR, MCWHORTER RL JR. ESTES EH JR. WARRENJV, MERRILL AJ: The effect of application of tourniquets to the legson cardiac output and renal function in normal human subjects. JC/in Invest 32:1163—1 170, 1953

23. FELDMAN D, FUNDER J, EDELMAN I: Subcellular mechanisms inthe action of adrenal steroids. Am J Med 53:545—560, 1972

24. SCHULTZE RG, SHAPIRO HS, BRICKER NS: Studies on the controlof sodium excretion in experimental uremia. J C/in In vest 48:869—877, 1969

25. SCHRIER RW, REGAL EM: Influence of aldosterone on sodium,water and potassium metabolism in chronic renal disease. KidneyInt 1:156—168, 1972

26. Hormonal regulation of sodium excretion. Proceedings of thesatellite symposium of the 28th I.V.P.S. Congress, edited byLICHARDUS B, SCHRIER RW, PONEC J, Amsterdam, Elsevier/NorthHolland, 1980

27. BOURGOIGNIE JJ, HWANG KH, ESPINEL C, KLAHR 5, BRICKERNS: A natriuretic factor in serum of patients with chronic uremia. JC/in Invest 51:1514—1527, 1972

28. BOURGOIGNIE JJ, HWANG KH, IPAKCHI E, BRICKER NS: Thepresence of a natriuretic factor in urine of patients with chronicuremia. The absence of the factor in nephrotic uremic patients. JC/in Invest 53:1559—1567, 1974

29. BROWN PR, K0UTSAIMANIS KG, DE WARDENER HE: Effect ofurinary extracts from salt-loaded man on urinary sodium excretionby the rat. Kidney Int 2:1—5, 1972

30. FAVRE H, HWANG KH, SCHMIDT RW, BRICKER NS,BOURGOIGNIE JJ: An inhibitor of sodium transport in the urine of

dogs with normal renal function. J C/in Invest 56:1302—1311, 197531. EPSTEIN M, BRICKER NS, BOURGOIGNIE JJ; The presence of a

natriuretic factor in urine of normal men undergoing water immer-

sion. Kidney In! 13:152—158, 197832. VANLANTHAN M, Louis F, FAVRE H: Importance relative du

facteur natriuretique et de l'aldosterone dans le maintien du bilan

sodique. J d'Urolog Nephro/ 85:569—573, 197933. KLAHR 5, RODRIGUEZ HJ: Natriuretic hormone. Nephron 15:387—

408, 197534. SCHMIDTRW, BOURGOIGNIE JJ, BRICKER NS: On the adaptation in

sodium excretion in chronic uremia: The effects of "proportionalreduction" of sodium intake. J Clin Invest 53:1736—1741, 1974

35. FINE LG, BOURGOIGNIE JJ, HWANG KH, BRICKER NS: On theinfluence of the natriuretic factor from patients with chronic uremiaon the bioelectric properties and sodium transport of the isolatedmammalian collecting tubule. J C/in Invest 58:590—597, 1976

36. CLARKSON EM, RAW SM, DE WARDENER HE: Further observa-tions on a low-molecular-weight natriuretic substance in urine ofnormal man. Kidney mt 16:710—721, 1979

37. BRICKER NS, LICHT A: Natriuretic hormone: Biologic effects andprogress in identification and isolation, in Proceedings of Sympo-sium on Hormona/ Regu/ation of Sodium Excretion, edited byLICHARDUS B, SCHRIER RW, PONEC J,Amsterdam, Elsevier/NorthHolland, 1980, pp. 399—408

38. KAPLAN MA, B0uRG0IGNiE JJ, ROSECAN J, BRiCKER NS: Theeffects of the natriuretic factor from patients with chronic uremia onthe bioelectric properties and sodium transport of the isolatedmammalian collecting tubule. J C/in Invest 58:590—597, 1976

39. BRICKER NS, LICHT A: Natriuretic hormone: Current status, inTextbook of Nephrology, edited by MASSRY SO, GLASSOCK RJ,New York, Elsevier/North Holland, in press

40. GRUBER KA, WHITAKER JM, BUCKALEW WM JR: Endogenousdigitalis-like substance in plasma of volume-expanded dogs. Nature287:743—745, 1980

41. FINE LG, BOURGOIGNIE JJ, WEBER H, BRICKER NS: Enhancedend-organ responsiveness of the uremic kidney to the natriureticfactor. Kidney mt 10:364—372, 1976

42. WEBER H, B0uRGOIGNIE JJ, BRICKER NS: Effects of the natriureticserum fraction of proximal tubular sodium reabsorption. AmJPhysiol 226:419—425, 1974

43. DANOVITCH GM, JACOBSON E, LICHT A: Absence of renal sodiumadaptation in chronic renal failure. Am J Nephrol, in press

44. FINE LG, YANAGAWA N, SCHULTZE RG, TUCK M: Functionalprofile of the isolated uremic nephron: Potassium adaptation in the

rabbit cortical collecting tubule. J Clin Invest 64:1033—1043, 197945. SLATOPOLSKY E, CAGLAR S, PENNELL JP, TAGGART DD, CANTER-

BURY JM, REISS E, BRICKER NS: On the pathogenesis of hyper-parathyroidism in chronic renal insufficiency in the dog. J C/in

Invest 50:492—499, 197146. ALT5HELER P, KLAHR 5, ROSENBAUM R, SLATOPOLSKY E: Effects

of inhibitors of prostaglandin synthesis on the renal excretion ofsodium in normal dogs and dogs with decreased renal mass. Am JPhysio/ 235:F338—F344, 1978

sodium homeostasis in chronic renal disease · sodium homeostasis in chronic renal disease...

Documents