Erythropoiesis in Anephric Man - JCI

Journal of Clinical InvestigationVol. 43, No. 11, 1964

Erythropoiesis in Anephric Man*DAVID G. NATHAN, EUGENESCHUPAK, FREDERICK STOHLMAN, JR., AND

JOHNP. MERRILL(From the Departments of Medicine of the Peter Bent Brigham Hospital and Harvard Medical

School and of St. Elizabeth's Hospital and Tufts University School ofMedicine, Boston, M1ass.)

Recent progress in the understanding of thephysiology of erythropoiesis has led to definitionof the kidney's importance as a source of erythro-poietin. Although Erslev (1, 2) did not docu-ment erythropoietin production in rabbit kidneyin vivo, Jacobson, Goldwasser, Fried, and Plzak(3), Reissmann, Nomura, Gunn, and Brosius (4),and Naets and Heuse (5, 6) clearly managed todefine the kidney as an important regulator oferythropoiesis in rats and dogs. The studies inthe dog were very convincing, for Naets (5) de-scribed nearly total and permanent erythroidaplasia after bilateral nephrectomy.

Increasing ability to manage chronic renal fail-ure and the renoprival state, by both renal trans-plantation and intermittent, long-term hemodialy-sis, has offered the opportunity to examine thepossible role of the kidney in human erythropoie-sis. Several uremic patients have undergone bi-lateral nephrectomy, remaining renoprival forfrom 2 to over 400 days before suitable homotrans-plantation could be performed. One individual,in whom a solitary kidney was accidentally re-moved, was subsequently transferred to the PeterBent Brigham Hospital and studied before trans-plantation. In addition, erythropoiesis was meas-ured in two patients who were considered for, butdid not undergo, bilateral nephrectomy. The

* Submitted for publication April 7, 1964; acceptedJuly 13, 1964.

Supported by U. S. Public Health Service grantsAM-00965-09, HE-07542-IR-01, and HE-08260-01; U. S.Army contract DA 49193 MD 2497; and the John A.Hartford Foundation. Some of these clinical studieswere performed in the Clinical Research Center ofHarvard Medical School and Peter Bent Brigham Hos-pital, supported by grant 8-MO1-FR-31-04 of the NationalInstitutes of Health.

Presented in part at the Sixth Annual Meeting of theAmerican Society of Hematology, December 9-10, 1963,Washington, D. C.

erythropoietic status of these uremic and anephricindividuals forms the basis of this report, the re-sults of which fail to demonstrate complete de-pendence of erythropoiesis on the presence ofrenal tissue in mnan.

Methods

1) Patients. A total of six patients was investigated(Table I). All were hospitalized at the Peter BentBrigham Hospital. Four patients were maintained inthe renoprival state for periods of 2 to 420 days. Pa-tients 1 and 2 had bilateral nephrectomy in preparationfor renal transplantation. In the former, nephrectomywas carried out in separate stages. Patient 3 underwentbilateral nephrectomy in separate stages because of ac-celerated, unrelenting hypertension associated with chronicpyelonephritis. Patient 4, previously nonazotemic, wasaccidentally rendered renoprival by removal of a soli-tary kidney. He was transferred to the Peter BentBrigham for a renal homograft. Patients 5 and 6 werestudied with both kidneys in situ, while being maintainedon a chronic hemodialysis program.

2) Treatment. Hemodialysis, required by all patientswith the exception of Patient 1, who was maintained byperitoneal lavage, was generally carried out twice weekly.Patient 3 was hemodialyzed for several hours five morn-ings each week. Dialysis was performed utilizing theTravenol twin coil artificial kidney. Frequently onlyone coil was used, eliminating the need for exogenousblood. The number of hours of dialysis and the bathcomposition were dependent upon the individual patient'sneed. Dietary salt, water, and protein were individuallyvaried. In general, moderate protein and sodium re-strictions were imposed, and all patients kept theirfluid intake below 1 L per day.

3) Hematologic investigations. Hematopoiesis wasinvestigated during the anephric state in four of thepatients. The diseased kidneys of Patients 5 and 6 re-mained in situt. Hematocrit, hemoglobin concentration,reticulocyte percentage, leukocyte count, and plateletcount (phase) were measured with standard techniques(7). Plasma iron turnover and red cell production rateswere estimated with Fe59 by techniques previously de-scribed (8). In vivo external scanning studies of Fe5"localization in liver, spleen, and sacrum, and Cr" redcell survival studies were performed in four patients.

2158

ERYTHROPOIESISIN ANEPHRICMAN

FIG. 1. SELECTED OIL IMMERSION VIEWS OF THE BONE MARROWASPIRATES FROM PATIENT 5,WHOSEDISEASED KIDNEYS REMAINED IN SITU; PATIENT 3, 28 DAYS AFTER NEPHRECTOMY;AND

PATIENT 4, 11 (left) AND 131 (right) DAYS AFTER NEPHRECTOMY. Erythroid precursors are

easily demonstrable.

The time betwen hemodialyses was never greater than7 days; therefore the Cr' survival measurements can-

not be considered accurate.Bone marrow aspirates were stained with the Jenner-

Giemsa technique as well as with ferrocyanide, the latterfor estimate of iron stores.

4) Effects of hypoxia. Attempts to provide controlledhypoxic stimulation of erythropoiesis in Patients 3 and4, by exposure to a gas mixture of low oxygen tension,were unsuccessful. The patients were unable to toleratethe confinement necessary to complete these studies.Hypoxia was induced in Patient 3 by an intercurrentillness, and estimates of erythropoiesis were performedbefore, during, and after this event (see Results).

5) Erythropoietin assays. Plasma erythropoietin levelswere estimated both at rest and before and after the in-duction of anoxia. Alcoholic extracts (9) of 100-mlsamples of plasma were assayed in polycythemic mice(10). Both Fe' uptake by circulating erythrocytes andthe reticulocyte responses were measured. In neitherbioassay system were the erythropoietic activities of theseplasma extracts greater than those of control plasma ex-

tracts or of saline.

Results

1) The peripheral blood counts of the patientsat the time of their initial bone marrow and iso-tope studies are shown in Table II.

TABLE I

Azotemia and kidney weight

Duration ofTotal kidney azotemia be-

Renoprival Blood urea weight fore initialPatient Age Sex Disease period nitrogen study

days mg/100 ml g1 36 M Glomerulonephritis 2 112 240 8 mo2 17 F Glomerulonephritis 30 40-100 130 7 yr3 35 M Pyelonephritis 420 30-70 200 20 yr4 34 M No prior renal disease 145 25-175 10 days5 29 M Pyelonephritis 50-125 125 (autopsy) 3 yr6 45 M Glomerulonephritis 50-125 6 yr

2159

NATHAN, SCHUPAK, STOHLMAN,AND MERRILL

PLASMA FE59 CLEARANCE

RB.C. FE59 UPTAKE

DAYS

FIG. 2. THE PLASMA FE' CLEARANCEAND RED CELL

FEU UPTAKE CURVES IN THE ANEPHRIC AND UREMIC PA-

TIENTS. Heavy lines denote the average normal valuesusually observed.

2) Cytologic examinations of bone marrow fromthese patients were not of predictive value whenrelated to measurements of effective red cell pro-

duction with Fe59. Some selected oil immersionfields from Patients 3, 4, and 5 are shown inFigure 1. Contrary to results in dogs (5, 6, 11),erythroid cells were apparent in all of the prepa-

rations, but the marrow smears of these patientstended to be hypoplastic. The marrow iron

stores were uniformly increased.

3) Table III presents the data derived fromthe studies with Fe59 and Cr51. Figure 2 portrays

the plasma Fe59 clearance and red cell Fe59 uptakecurves. Figure 3 shows the results of in vivoscanning of liver, spleen, and sacrum in Patients2, 3, 4, and 5. These data are in no way distinc-tive from data derived from similar examinationsof uremic individuals (12). Erythropoiesis was

suppressed, but the degree of suppression of eryth-rocyte production was certainly no more strikingthan that observed in patients with chronic renaldisease. In fact, Patient 5, whose kidneys re-

mained in situ, had one of the poorest red cellproduction rates of the group. In addition, Pa-tient 4, who had been uremic for only 10 days,had an increased rate of erythropoiesis that failedto compensate for his short erythrocyte survival.It should be emphasized that the patients had lostvarying amounts of their body fat; this alterationin body composition might result in some over-

estimation of the red cell production rate ex-

pressed as milliliters per kilogram per day.The results of serial studies of Patient 3 showed

that this anephric patient could enhance his rateof erythropoiesis. Two attempts were made toexamine the effects of hypoxia in this patient.The first, on days 119 and 120 after, nephrectomy,was a controlled study that was poorly toleratedby the patient. Hypoxia was intermittent. Noeffect on plasma or red cell iron turnover couldbe demonstrated, nor were plasma erythropoietinlevels affected. On day 329, 1 week before thenext measurement, the patient underwent subtotalgastrectomy because of persistent hemorrhagefrom a duodenal ulcer. He received 16 U of blood.On day 334 he became progressively hyperpneicand dyspneic; his arterial Po2 and hemoglobin

TABLE II

Hematologic data

Dayspost- Blood

nephrec- Hema- Hemo- Reticulo- ureaPatient tomy tocrit globin Leukocytes Platelets cytes nitrogen

% g/100 ml /mms /mm' % mg/100 ml1 2 28 8.7 23,075 1.0 912 16 25 7.5 8,562 112 -000 0.8 243 27 27 9.2 1 1 ,400 232,500 2.2 894 9 21 6.7 13,523 3.0 1665 19 5.7 6,540 240,000 0.5 916 22 7.0 4,344 345,000 1.0 104

2160

ERYTHROPOIESISIN ANEPHRICMAN

0000 el;0%00 00

eto NOU) t.

U) U)

e, '0 U') U) U_ -4tN0 00 0 o t-N N N q CN C eq1Nm qNN

ei --_

-000~d ~o N t

"

Uo o00 U-) oO U

; _. e-

U.-

66666_o 0 t0

-t 00 00 ~~0% 00

1-1 NoWe

U--~~~~~-

'o No to o It in

%0)U %QUxo U_ 'U) o

4o 00 U) 00UU)O0oZt- r 0_ ON

- N N -___ -

U)

U) %0 '0

0 U)

0o.

U) U)O -4 006 1.- 6

0

eU U) oU0 6 oo

0

U)oo

U)

0o m'0 U)

1-4 U-)

e6 Ul

0 No o C- o 0oo _ -tn\ 4u4oO '0 U)coo -o 66 6 6 6 6

0U)

%)oU) 0%0NO 4 .

WU _ %ONO - 00o NONO

NO' N_- _-

(N '0 0o o0 0U t o-_-

(N U)No _ r" C1 m m _ 4

Cd-4 Ul) No E

0

Cdla

*Ni

0 C

(U.8

2161

co

10P4

0- 4

o4

o 6

4).4) s

4 U)N.O_0 "

U)

8 N s0 %

t - .

4)

800)N.

0._

4)

0

0

010

4)- 'A

r. >0 00

00 (U

00i

bG >-wQr.!

i5co

dU)~.0 E0 0

0

(U

NATHAN, SCHUPAK, STOHLMAN,AND MERRILL

BODY SURFACE SCANNING-FE59

C.PM.

DAYS

FIG. 3. BODYSURFACESCANNINGOF FE' RADIOACTIVITY OVERTHE LIVER, SPLEEN, AND SACRUMOFPATIENTS 2, 3, 4, AND 5. COUNTSARE CORRECTEDFOR THE RADIOACTIVITY DUE TO BLOODFLOW.

oxygen saturation were at the low levels recordedin Table III. The cause of this disorder wasnot determined. The measurements of erythro-poiesis on day 336 were performed approximately30 hours after onset of these symptoms and re-vealed a marked increase in red cell productionover previous recorded values. No erythropoietinactivity was detected in his plasma. It is im-portant to stress that the high red cell volumemeasured on day 336 resulted from transfusiontherapy. In addition, the higher estimate of redcell production resulted from the increased plasmavolume, which was associated with a noteworthyincrease in the serum iron concentration and afaster rate of disappearance of Fe59 from theplasma. The hypoxia gradually improved. Onday 363, after 2 weeks of freedom from respiratorysymptoms, erythropoiesis was studied once again.The red cell production had declined somewhatbut not to the levels previously measured in thispatient. On renoprival day 398, this patient de-

veloped severe pericardial tamponade due to abloody uremic pericardial effusion. This hypoxicstimulus persisted for 24 hours, until it was re-lieved by partial pericardectomy. Reticulocytecounts of 8 to 10% were observed during the firsttwo postoperative days.

Examination of red cell production was re-peated in Patient 4 on renoprival day 131. Thisstudy was accompanied by a bone marrow exami-nation and was carried out 2 weeks after the pa-tient's last transfusion. In addition, he had beendialyzed for the previous 2 weeks on the artificialkidney without bank blood priming of the coil.Since the turnover time of exogenously adminis-tered erythropoietin is thought to be no greaterthan 1 to 2 days (13) and may be much morerapid (9), it is very unlikely that any exogenouserythropoietin was available to the patient duringthis study. The bone marrow was somewhat hy-poplastic, but the myelopoietic: erythropoietic ra-

tion was 3: 1, and the measurement of red cell

2162

ERYTHROPOIESIS IN ANEPHRICMAN

production with Fe59 revealed a slightly increasedrate of red cell production.

The results of the bone marrow and ferrokineticmeasurements in Patient 6, whose kidneys re-

mained in situ, were similar to those of theanephric patients.

Discussion

The kidney appears to be an important source

of erythropoietin in dogs, rabbits, and rats (3-6,11, 14). Goldfarb and Tobian (15), who havereviewed the rodent juxtaglomerular apparatusas a source of erythropoietin, postulate the exist-ence of at least two rodent renal erythropoietins.Erythropoietic function has been ascribed (16)and denied (17) to angiotensin II. The growingliterature describing erythrocytosis in associationwith renal cysts and tumors (18-20) also sup-

ports the importance of the kidney as a source

of this hormone. On the contrary, reports con-

cerning erythrocytosis with or without increasederythropoietin production in association with cere-

bellar and hepatic tumors in man (21, 22) indi-cate that extrarenal neoplastic tissues may alsoinfluence erythropoiesis. Erslev's studies of therabbit (1, 2) firmly demonstrate lack of depen-dence of erythropoiesis on the kidney in this ani-mal. In addition, studies of erythropoietic ac-

tivities in non-neoplastic extrarenal tissue extracts(23, 24) and of erythropoietic responses inanephric rabbits and rats (24, 25) have been in-terpreted to indicate a less restricted site of eryth-ropoietin synthesis and broader control of eryth-ropoiesis. Indeed, recent evidence indicates thatcertain products of red cell hemolysis may them-selves contribute to stimulation of erythropoiesis(26).

None of our patients had detectable erythro-poietin in extracts of their plasma. However,the development of more sensitive methods mightreveal its presence. Gallagher, McCarthy, andLange (27) and Naets and Heuse (6) have alsonoted that anemic uremics usually have no eryth-ropoietin detectable in their plasma. Althoughthese workers interpret their data to support theview that "the erythroid failure of uremia is dueto inability of the diseased kidney to produceerythropoietin" (27), it is clear that many otherplasma inhibitors created in the uremic syndrome

may combine to reduce the detectable erythro-poietin in a crude plasma extract. In addition,serious interference with the morphology of HeLa(28) and bone marrow cells (29), as well aschemical interference with erythrocyte metabolismin uremia (30-32), has been described.

Although Choremis, Megas, Liaromati, andMichael (33) maintain that marrow erythroblas-topenia occurs in cases of acute glomerulonephri-tis without azotemia, review of their cases revealsonly two such patients, and one of these was onlymildly anemic. Moreover, the relationship be-tween azotemia and erythropoiesis is highly vari-able. Some patients are observed with acuterenal cortical necrosis, marked azotemia, and he-molytic anemia with reticulocyte counts as highas 15 to 30%. Others with mild azotemia areanemic due to depressed erythropoiesis. In ad-dition, Van Dyke, Keighley. and Lawrence (34)have studied one patient with mild anemia sec-ondary to chronic glomnerulonephritis. This pa-tient failed to exhibit an erythropoietic responseto a 3-day course of erythropoietin at a dose thatwas considered large enough to produce a sig-nificant erythropoietic response in normal indi-viduals.

The results of study of the anephric men de-scribed here, and of a similar study performedseveral years ago by Rees and his co-workers(35), support a broad view of the regulation oferythropoiesis in man. The Cr5' measurements,although brief, agreed with previous findings ofhemolysis in uremia and particularly in the reno-prival state (36). The Fe59 studies showed thatthe anephric patients, in common with other ure-mics, all exhibited relative bone marrow failurein that they did not maintain an adequate mar-row response to the shortened erythrocyte sur-vival and low total red cell volume. But it mustbe emphasized that the anephric patients producedred cells as well as do uremic individuals whosekidneys remain in situ. Furthermore, anephricman may increase his red cell production ratewell above normal, as demonstrated by Patients3 and 4. These facts provide strong evidenceagainst complete dependence on renal tissue forred cell production in man. It should also benoted that the anephric patients maintained ade-quate circulating leukocytes and platelets.

2163

NATHAN, SCHUPAK, STOHLMAN,AND MERRILL

Our studies do not settle the question of thesource of erythropoietin, since the hormone wasnot detected at any time. If erythropoietin isnecessary for any erythropoiesis in man, the kid-ney is certainly not its sole source. On the otherhand, erythropoietin may not be necessary for thestimulation of the rates of erythropoiesis observedin these patients.

The studies of anephric patients with long his-tories of chronic uremia must be interpretedcautiously. The fact that in these patients erythro-poiesis was no worse than in severe chronic ure-mics, such as Patients 5 and 6, should be evalu-ated in the light of the physiological significanceof nephrectomy in such patients. A review of thetotal kidney weight (Table I) and renal morphol-ogy of the chronic uremics and of the renoprivalchronic uremics reveals that the surgeon usuallyremoved mere remnants of renal tissue. Theseanephric patients had performed a gradual andpartial autonephrectorny during the years beforesurgery. If the kidney is a prime source oferythropoietin, whatever secondary sources oferythropoietin might have been available wereallowed adequate time to rise. On the contrary,Patient 4 and the patient studied by Rees and hiscolleagues (35) had been entirely normal beforenephrectomy. The increased rate of erythropoiesisobserved shortly after nephrectomy in these pa-tients militates more strongly against an exclusiverenal role than do the results in the other patients.

Summary

Erythropoiesis has been measured in four reno-prival patients. These patients maintained hemo-globin concentrations within the range of 7 to 9 gper 100 ml of blood in the presence of mild he-molysis. Further, the response of one patient tohypoxia, although diminished, persisted in the ab-sence of the kidney. In another patient, an in-creased rate of red cell production was noted.It follows that man is not totally dependent upona renal source of erythropoietin for his red cellproduction.

AcknowledgmentsThe authors are grateful for the advice and assistance

of Drs. John Parker-Williams, Frederick Morgan, JohnM. Kinney, and Frank H. Gardner. The assistance of

former Harvard medical students, Drs. Dale Cowan andJohn Mueller, is gratefully acknowledged.

References1. Erslev, A. J. Erythropoietic function in uremic rab-

bits. Arch. intern. Med. 1958, 101, 407.2. Erslev, A. J. Erythropoietic function in uremic rab-

bits. II. Effect of nephrectomy on red cell produc-tion and iron metabolism. Acta Haemat. (Basel)1960, 23, 226.

3. Jacobson, L. O., E. Goldwasser, W. Fried, and L.Plzak. Role of the kidney in erythropoiesis.Nature (Lond.) 1957, 179, 633.

4. Reissmann, K. R., T. Nomura, R. W. Gunn, andF. Brosius. Erythropoietic response to anemia orerythropoietin inj ection in uremic rats with orwithout functioning renal tissue. Blood 1960, 16,1411.

5. Naets, J. P. Le r6le du rein dans erythropoiese.Acta clin. belg. 1960, 15, 359.

6. Naets, J. P., and A. F. Heuse. Measurement oferythropoietic stimulating factor in anemic pa-tients with or without renal disease. J. Lab. clin.Med. 1962, 60, 365.

7. Page, L. B., and P. J. Culver, Eds. A Syllabus ofLaboratory Examinations in Clinical Diagnosis.Cambridge, Harvard University Press, 1960.

8. Nathan, D. G., and N. I. Berlin. Studies of the pro-duction and life span of erythrocytes in myeloidmetaplasia. Blood 1958, 14, 668.

9. Stohlman, F., Jr., and D. Howard. Humoral regu-lation of erythropoiesis. IX. The rate of disap-pearance of erythropoietin from the plasma inErythropoiesis, L. 0. Jacobson and M. Doyle, Eds.New York, Grune & Stratton, 1962, p. 120.

10. Gurney, C. W., R. Degowin, D. Hof stra, and J.Byron. Applications of erythropoietin to biologi-cal investigations in Erythropoiesis, L. 0. Jacob-son and M. Doyle, Eds. New York, Grune &Stratton, 1962, p. 151.

11. Naets, J.-P. The role of the kidney in erythropoie-sis. J. clin. Invest. 1960, 39, 102.

12. Kaye, M. The anemia associated with renal disease.J. Lab. clin. Med. 1958, 52, 83.

13. Hammond, G. D., and A. Ishikawa. The rate ofdisappearance of erythropoietin following trans-fusion of severely anemic patients in Erythropoie-sis, L. 0. Jacobson and M. Doyle, Eds. NewYork, Grune & Stratton, 1962, p. 128.

14. Kuratowska, Z., B. Lewartowski, and E. Michalak.Studies on the production of erythropoietin by iso-lated perfused organs. Blood 1961, 18, 527.

15. Goldfarb, B., and L. Tobian. The interrelationshipof hypoxia, erythropoietin and the renal j uxta-glomerular cell. Proc. Soc. exp. Biol. (N. Y.)1962, 111, 510.

2164

ERYTHROPOIESISIN ANEPHRICMAN

16. Fisher, J. W., and J. J. Crook. Influence of severalhormones on erythropoiesis and oxygen consump-tion in the hypophysectomized rat. Blood 1962,19, 557.

17. Bilsel, Y. C., J. E. Wood, and R. D. Lange.Angiotensin II and erythropoiesis. Proc. Soc. exp.Biol. (N. Y.) 1963, 114, 475.

18. Gardner, F. H., and J. G. Freymann. Erythrocy-themia (polycythemia) and hydronephrosis. NewEngl. J. Med. 1958, 259, 323.

19. Rosse, W. F., T. A. Waldmann, and P. Cohen.Renal cysts, erythropoietin and polycythemia.Amer. J. Med. 1963, 34, 76.

20. Donati, R. M., R. D. Lange, and N. I. Gallagher.Nephrogenic erythrocytosis. Arch. intern. Med.1963, 112, 960.

21. Waldmann, T. A., and W. F. Rosse. Sites of for-mation of erythropoietin in Erythropoiesis, L. 0.Jacobson and M. Doyle, Eds. New York, Grune& Stratton, 1962, p. 87.

22. Kan, Y. W., A. J. S. McFadzean, D. Todd, andS. C. Tso. Further observations on polycythemiain hepatocellular carcinoma. Blood 1961, 18, 592.

23. Rambach, W. A., H. L. Alt, and J. A. D. Cooper.Erythropoietic activity of tissue homogenates.Proc. Soc. exp. Biol. (N. Y.) 1961, 108, 793.

24. Reissmann, K. R., and T. Nomura. Erythropoietinformation in isolated kidneys and liver in Erythro-poiesis, L. 0. Jacobson and M. Doyle, Eds. NewYork, Grune & Stratton, 1962, p. 71.

25. Rosse, W. F., and T. A. Waldmann. The role of thekidney in the erythropoietic response to hypoxiain parabiotic rats. Blood 1962, 19, 75.

26. Brown, J. R., N. A. Altschuler, and J. A. D. Cooper.Erythropoietic effect of red blood cell componentsand heme-related compounds. Proc. Soc. exp. Biol.(N. Y.) 1963, 112, 840.

27. Gallagher, N. I., J. M. McCarthy, and R. D. Lange.Observations on erythropoietic-stimulating factor(ESF) in the plasma of uremic and nonuremicanemic patients. Ann. intern. Med. 1960, 52, 1201.

28. Henkin, R. E., P. H. Byatt, and M. H. Maxwell.Evidence for the presence of a dialyzable "toxic"factor in the sera of uremic patients. Clin. Res.1961, 9, 202.

29. Berman, L., and E. R. Powsner. Review of meth-ods for studying maturation of human erythro-blasts in vitro: evaluation of a new method of cul-ture of cell suspensions in a clot-free medium.Blood 1959, 14, 1194.

30. Morgan, J. M., and R. E. Morgan. Study of theeffect of the uremic metabolites on erythrocyteglycolysis. Metabolism 1964, 13, 629.

31. Morgan, J. M., R. E. Morgan, and G. E. Thomas.Inhibition of lactic dehydrogenase by ultrafiltrateof uremic blood. Metabolism 1963, 12, 1051.

32. Bock, H. E., G. W. Lohr, and H. D. Waller.Beitrag zur Pathogenese renaler Anamien.Schweiz. med. Wschr. 1962, 92, 1213.

33. Choremis, C., H. Megas, A. Liaromati, and S.Michael. Bone marrow studies in acute glomeru-lonephritis. Helv. paediat. Acta 1962, 17, 138.

34. Van Dyke, D., G. Keighley, and J. Lawrence.Decreased responsiveness to erythropoietin in apatient with anemia secondary to chronic uremia(abstract). Blood 1963, 22, 838.

35. Rees, S. B., W. G. Scheitlin, C. Giordano, W. R.Guild, and J. P. Merrill. Pathologic physiologyof the anemia associated with renal failure (ab-stract). Excerpta med. (Amst.) Series XXIX.First Int. Congr. Nephrology 1961, 1, 67.

36. Muirhead, E. E., and F. Jones. Renoprival hemoly-sis and its prevention by renal tissue. Blood 1963,22, 272.

2165