SUSCEPTIBILITY TO EXPERIMENTALPYELONEPHRITIS WHENHORMONALHYPERTENSIONIS PREVENTEDBY

HYPOTEN§IVEDRUGS*

By JAMESW. WOODS

(From the Department of Medicine, University of North Carolina School of Medicine,Chapel Hill, N. C.)

(Submitted for publication April 29, 1960; accepted August 3, 1960)

Rats with desoxycorticosterone (DCA) -salineinduced hypertension have been shown to exhibitincreased susceptibility to experimental pyelo-nephritis (1, 2). These studies lend support tothe hypothesis that various injuries, includingnephrosclerosis, predispose to human pyelonephri-tis and that pyelonephritis is superimposed uponhypertensive disease more frequently than is usu-ally suspected. Others (3-5) have shown thatreserpine and/or hydralazine, when given to ratswith DCA-saline hypertension, will reduce bloodpressure, delay or prevent renal and vascularpathological changes, and increase the averagesurvival time. Masson, McCormack, Dustan andCorcoran (6) demonstrated a close associationbetween blood pressure levels and the presenceand character of the vascular lesions in rats withhydralazine-treated renal hypertension. The pre-viously demonstrated susceptibility of rats withDCA-saline hypertension might have been dueeither to an effect of the hypertension and vasculardisease or to some other effect of the administeredsteroid and saline. The present investigation wastherefore designed to separate these possibilities.

MATERIALS AND METHODS

White, female Sprague-Dawley strain rats, weighing100 to 150 g, were used. The experimental animals ofGroups A and B were subjected to left nephrectomy and3 fortnightly injections of a long-acting derivative ofDCA1 in 25 mg doses. Drinking water contained 1 percent NaCl and 0.49 per cent KCI. The latter had beenpreviously found (1) to prevent potassium depletionwhich otherwise results from the DCA-saline regimen.In addition, 40 mg of hydralazine' was added to eachliter of the drinking solution of Group A animals andthey received 1 mg per kg (later reduced to 0.5 mg per

* This investigation was supported by a grant from TheLife Insurance Medical Research Fund.

1 Kindly supplied by Ciba Pharmaceutical Products,Inc., Summit, N. J.

kg) of syrosingopine1 subcutaneously each day. Syro-singopine was chosen rather than reserpine since it hasbeen found to have little, if any, effect on spontaneoussalt appetite (7).

The animals of Group C had a left nephrectomy butreceived no DCA and drank ordinary tap water. Allanimals ate Purina Laboratory Chow. The animals were"pair-fed" by groups, i.e., the intakes of Groups A and Cwere limited each day to the average intake of Group B.Additionally, the fluid intake of Group B animals waslimited each day to the average fluid intake of Group Ain order to adjust for any effect syrosingopine mighthave on spontaneous salt intake. The resulting weightcurves of the three groups were very similar, as isshown in Figure 1.

The method of determining blood pressure by use ofa microphonic manometer and the bacteriological tech-niques employed have been previously described (1, 8).Rats of the three groups designated for bacterial injec-tion received an inoculum of 100,000,000 Escherichia colii.v.

Experimental design. The experimental design isshown in Table I. Administration of hypotensive drugsand the DCA-saline regimen were started simultaneously.

Intravenous injection of E. coli was carried out 7 weekslater and the rats sacrificed 2 weeks following injection.Six rats of Groups A and B and 5 of Group C did notreceive E. coli, in order to facilitate histological study of

WEEKS

FIG. 1. AVERAGEWEIGHTSAND SYSTOLIC BLOODPRESSURESOF GROUPSA, B AND C.

1813

JAMES W. WOODS

TABLE I

Experimental design

-No. of rats No. of ratsreceiving uninjecte(1

Groul) Description E. coli i.v. with E. coli

A DCA-saline regimiieni 15 6uninephrectoimiysyrosingopine-hydralazine

B DCA-saline regimiieni 16 6tininephrectomyn1o hypotensive (Irtugs

C ITninephrectomnized conitrols 14 5}......i4.a Sb . . . . ........ozR:y-,t ;- a W ",=i*

FIG. 2. ILLUSTRATIONS OF NORMALAND PATHOLOGICAL

HISTOLOGYENCOUNTERED. 1. Kidney of control rat (previ-ously uninephrectomized). 2. Kidney from an uninjectedrat of Group A showing a nodular deposit of material

the kidneys of these representatives after sacrifice. Atthe time of sacrifice, the remaining right kidney of eachanimal was removed with aseptic technique, divided inhalf longitudinially, and one-half cultured quantitativelywhile the other half w/as fixed in 10 per cent formalin forhistological sections.

RESULTS

General condition of animiials. While equal num-bers (nine) of rats (lied in Groups A and B priorto injection of E. coli there appeared to be a dif-ference between the two groups. All of the GroupA rats were markedly edematous at the time ofdeath and six of nine died within the first eightdays of the experiment. This, correctly or not,was attributed to the syrosingopine, and the mor-tality rate and development of edema seemed todecline markedly when the dose of this drug wasreduced from 1 to 0.5 mg per kg per day. Deathsof the nine rats of Group B were fairly evenlydistributed over the interval of 6 to 20 days afteronset of the experiment and all but one were thinand sickly in appearance. The one exception wasedematous. Four of the control animals in GroupC also died-one accidentally, one from intestinalobstruction, and two from unknown cause. Theremaining animals of the three groups maintainedbody weight, appeared reasonably healthy, and 'allsurvived to the time of sacrifice, two weeks fol-lowing intravenous injection of E. coli.

Systolic blood pressure. As is seen in Figure 1,the combination of syrosingopine and hydralazinewas highly successful in preventing severe hyper-

with the staining characteristics of fibrinoid in a glo-merulus and very slight tubular dilatation. 3. Kidneyfrom an uninjected rat of Group B showing periglomeru-lar fibrosis, glomerular sclerosis, marked dilatation oftubules, some of which contain casts, and minimal ar-teriolar sclerosis. Hematoxylin and eosin (X 200).

1814

MODIFIED HORMONALHYPERTENSIONAND PYELONEPHRITIS

tension. The average of the blood pressures ofthe Group A rats was practically identical withthat of the controls of Group C at the time ofbacterial injection. There were individual varia-tions in blood pressure levels, but no animal inthis group had a reading above 160 mmHg andmost had pressures below 130 mmduring thegreater part of the seven week period. The aver-

age blood pressure for Group B, on the otherhand, rose progressively to 200 mmHg by theend of seven weeks. The highest pressure re-

corded in this group was 236 mmHg.Renal histology. The rapid development of

hypertension in the albino rat as a result of theDCA-saline regimen is usually accompanied by theappearance of widespread acute vascular disease(5, 9). Arteriolar sclerosis is present and some

arterioles contain subendothelial material with thecharacteristics of fibrinoid. Glomeruli show en-

largement and capsular fibrosis. Convoluted tu-bules are often dilated and contain eosinophiliccolloid. The above described lesions were pres-

ent to a moderate or severe degree in the kidneysof rats of Group B. In most of the rats of GroupA, there were no or minimal glomerular and tu-bular changes, and fibrinoid arteriolar lesions were

rare. The two instances in this group of more

advanced renal changes also had the highest sys-

tolic blood pressure recorded. Illustrations of kid-neys of normal (uninephrectomized) rats, and ofthose of Groups A and B are shown in Figure 2.

Susceptibility to experimental pyelonephritis.The results are shown in Table II. Of the 15 ratsof Group A receiving one hundred million E. coliintravenously, 4 had unequivocal pyelonephritis.One had no histological evidence of infection, a col-ony count of only 800 bacteria per ml of renal tissue,and was considered uninfected. Of the 16 rats ofGroup B, 10 had clear-cut evidence of pyeloneph-ritis. One animal had both gross and microscopicevidence of abscess formation in the half-kidneywhich was fixed for sections, while the remaininghalf had no gross abscesses and was sterile on cul-ture. The null-hypothesis stating that the inci-dence of infection is the same for rats in Groups Aand B was rejected in favor of the alternative hy-pothesis that if a difference did in fact exist, itwould be in the direction of rats in Group B hav-ing a greater incidence of infection than rats inGroup A. This conclusion was arrived at afterapplication of the appropriate one-tailed test ofsignificance based upon Fisher's exact method fora 2 x 2 contingency table (10). The calculatedprobability of an error of the first kind (viz., thatof falsely concluding that the incidence of infec-tion for rats in Group B is greater than that forrats in Group A) is p = 0.049. Only one con-

trol of Group C had renal infection. This was evi-denced by a microscopic abscess and a colonycount of 9,500 organisms per ml of kidney. Thedifference in incidence of infection between GroupsB and C is highly significant (p = 0.002), as was

TABLE II

Colony counts and descriptions of kidneys of the three groups two weeks after injection of E. coli

Group A Group B Group C

Colony counts Description* Colony counts Description Colony counts Description

0 0 1,200,000 + + 0 03,700,000 + + + 8,000,000 + + + 0 0

800 0 11,000,000 + + + 0 05,000,000 + + + 2,000,000 + + + 0 0

20,000,000 +++ 2,100,000 + + 0 02,000,000 ++ 0 ++0+°

0 0 100,000 + + 0 00 0 1,400,000 + + + 0 00 0 18,000,000 + + 0 00 0 0 0 0 00 0 0 0 9,500 + +0 0 100,000,000 + + + 0 00 0 0 0 0 00 0 0 0 0 00 0 0 0

0 0

* 0, no cellular infiltrate; + +, microscopic abscesses; + + +, macroscopic abscesses.

1815

JAMES W. WOODS

previously found (1). There was no significantdifference in incidence of infection between GroupsA and C.

DISCUSSION

The renewal of interest in and great increasein research on pyelonephritis and unilateral renaldisease associated with hypertension in recentyears have raised many questions regarding theinterrelationships of these diseases in man. Fewwould question that chronic bilateral pyelonephri-tis may produce severe hypertensive disease, but itoften does not, even when leading to uremia.Much skepticism has arisen regarding the role ofunilateral pyelonephritis and hypertension, sinceremoval of the affected kidney so rarely results inalleviation of the hypertension. Even the produc-tion of hypertension in the experimental animal bymeans of pyelonephritis has not been firmly estab-lished. If such an experimental model can bedeveloped, it should be quite helpful in the eluci-dation of these problems.

The reverse relationship, i.e., that hypertensiverenal disease may predispose to pyelonephritis, isonly suggested so far by a few human and animalinvestigations (1, 2, 11, 12). Goldblatt (13) hashypothesized that malignant hypertension may de-velop from benign hypertension because of thesuperimposition of renal infection. Long-termstudies of human normal and hypertensive popula-tions will be required to settle these questions.

The data reported here again show that hyper-tensive disease induced by DCAand saline leadsto greatly increased susceptibility to experimentalpyelonephritis. They also suggest that the bloodpressure itself is an important factor in the pro-duction of renal damage and heightened suscepti-bility, since when it is controlled at normal ornear normal levels, renal arterial and arteriolarlesions are minimal or absent and susceptibility toinfection is reduced. That the blood pressure maynot be the only factor is demonstrated by the re-cent experiments of Gardner (5). He showedthat, in rats with steroid hypertension receivingdaily intramuscular injections of hydralazine,acute vascular disease can be prevented despitediscontinuous reduction in the mean systemic bloodpressure and suggested that this prevention wasdue either to the discontinuous effect on bloodpressure or to "a factor, at present unrecognized,

which is miiore sensitive to the action of the drugthan is the blood pressure." In this study, hy-dralazine was present in the drinking solution,syrosingopine was administered additionally, anda continuous hypotensive effect is assumed. How-ever, it cannot be proved, since blood pressuremeasurements were made at weekly intervals.

This study has made no attempt to investigatethe effect of hypotensive drugs on survival timeof the animals and, while the doses of syrosingo-pine used during the first week of the experimentin Group A seemed to cause edema and earlydeath, this is not certain. In this investigation,there was no significant difference in mortalityrate during the seven week period prior to bac-terial injection between the rats with controlledand untreated hormonal hypertension.

SUMMARY

Rats with desoxycorticosterone-saline inducedhypertension have again been found to exhibitmarked renal vascular damage and increased sus-ceptibility to experimental pyelonephritis. Inother rats on the same regimen to which hypo-tensive drugs were added, blood pressure was ator near normal levels when measured at weeklyintervals, renal vascular lesions were minimized,and no increased susceptibility to pyelonephritiswas demonstrated.

ACKNOWLEDGMENT

The author appreciates the technical assistance of MissBillie Sioux Bush and is indebted to Dr. Bernard Paster-nack, Department of Biostatistics, School of PublicHealth, for help with the statistical analysis and to Dr.William Huffines, Department of Pathology, for helpwith analysis of histological sections.

REFERENCES

1. Woods, J. W. Susceptibility of rats with hor-monal hypertension to experimental pyelonephri-tis. J. clin. Invest. 1958, 37, 1686.

2. Shapiro, A. P. Relationships of hypertension andrenal impairment to experimental chronic pyelo-nephritis in rats (abstract). J. clin. Invest. 1958,37, 930.

3. Gaunt, R., Antonchak, N., Miller, G. J., and Renzi,A. A. Effect of reserpine (Serpasil) and hydrala-zine (Apresoline) on experimental steroid hyper-tension. Amer. J. Physiol. 1955, 182, 63.

4. Gross, F., Noelpp, B., Sulser, F., Doebelin, R., andKiindig, H. Vergleichende Untersuchungen uberdie medikamentose Beeinflussung verschiedener

1816

MODIFIED HORMONALHYPERTENSIONAND PYELONEPHRITIS

Formen von experimenteller Hypertension. Klin.Wschr. 1955, 33, 372.

5. Gardner, D. L. The relationship between intermittenthypotension and the prevention by hydralazine ofacute vascular disease in rats with steroid hyper-tension. Brit. J. exp. Path. 1960, 41, 60.

6. Masson, G. M. C., McCormack, L. J., Dustan, H. P.,and Corcoran, A. C. Hypertensive vascular dis-ease as a consequence of increased arterial pres-

sure. Quantitative study in rats with hydrala-zine-treated renal hypertension. Amer. J. Path.1958, 34, 817.

7. Gaunt, R., Gross, F., Renzi, A. A., and Chart, J. J.Adrenal cortex in hypertension in Hypertension,J. H. Moyer, Ed. First Hahnemann Symp. on Hy-pertensive Disease. Philadelphia, Saunders, 1959,p. 225.

8. Woods, J. W., Welt, L. G., Hollander, W., Jr., andNewton, M. Susceptibility of rats to experimental

pyelonephritis following recovery from potassiumdepletion. J. clin. Invest. 1960, 39, 28.

9. Selye, H., Hall, C. E., and Rowley, E. M. Malig-nant hypertension produced by treatment withdesoxycorticosterone acetate and sodium chloride.Canad. med. Ass. J. 1943, 49, 88.

10. Fisher, R. A. Statistical Methods for ResearchWorkers, 12th ed. Edinburgh, Oliver and Boyd,1954.

11. Smythe, C. M., Rivers, C. F., and Rosemond, R. M.Incidence of infected urines in hypertensive sub-jects (abstract). Amer. J. Med. 1959, 27, 325.

12. Grieble, H. G., Johnston, L. C., and Jackson, G. G.A search for unsuspected pyelonephritis among

patients with hypertension (abstract). Clin. Res.1958, 6, 293.

13. Goldblatt, H. Pathogenesis of malignant hypertension(editorial). Circulation 1957, 16, 697.

1817