THE RATE OF REMOVALOF RADIOACTIVE IRON FROMTHEPLASMA-ANINDEX OF ERYTHROPOIESIS1

By LOUIS R. WASSERMAN,IRA A. RASHKOFF, DOROTHYLEAVITT, JOSE-PHINE MAYER, AND SHIRLEY PORT

(From the Hematology and Physics Laboratories and the Medical Services, The Mount SinaiHospital, New York, N. Y.)

(Submitted for publication August 1, 1951; accepted October 9, 1951)

Transport iron normally exists in the plasma asferric beta-l-globulinate. Since in most individualsthe greater part of the body iron is incorporated inthe hemoglobin of the red blood cells, it is probablethat changes in the production and destruction ofthese erythrocytes will be reflected in the turnoverof plasma iron. Radioactive iron of high specificactivity is particularly useful for a study of ironmetabolism since it can be injected intravenouslyin tracer amounts without changing the steadystate of the organism; the turnover of this radio-iron will then mirror the metabolism of the iron inthe body.

Flexner, Vosburgh, and Cowie (1) demon-strated that shortly after the intravenous injectionof radioiron into guinea pigs, all the tagged ironin the plasma was in combination with beta-l-glob-ulin. Following the injection of small quantitiesof such labelled plasma into other guinea pigs, thecirculating plasma radioiron level decreased ex-ponentially for the duration of the experiment.Labelled ferric beta-l-globulinate was also pro-duced in vitro by the above investigators by incu-bating radioactive ferric chloride with a smallamount of plasma. However, the intravenous in-jection of this mixture into guinea pigs was fol-lowed by a rapid loss of the plasma radioiron forabout 30 minutes, following which the decreaseagain became exponential. The early rapid losswas presumably due to incomplete combination ofthe iron with the beta-l-globulin of the plasma andthe diffusion of part of the injected iron into thetissues.

Similar results in humans utilizing in vitrotagged plasma were reported by Huff, Hennessyand Lawrence (2) and Huff and associates (3).Although in most instances, in vitro incubation ofradioiron with patients' plasma resulted in a con-

1 Aided by grants from the Anna Ruth Lowenberg andAlbert A. List Funds.

stant rate of disappearance of the radioactivityfrom the plasma following its injection intrave-nously, occasionally exponential disappearancerates were not obtained; the latter were thought tobe due either to insufficient metal combining pro-tein in the plasma or to the presence of an abnormalprotein.

The following investigations of the plasma irondisappearance rate were undertaken to elucidateone phase of iron metabolism in anemic individualsin whomthe anemia was due to bone marrow hy-pofunction; other patients with blood dyscrasiaswere also studied for comparison. A solution ofplasma fraction IV-7 2 was utilized for incubationwith the radioactive iron since it was found thatpatients' plasma was on the whole unsatisfactory,due to variations in the latent iron binding capaci-ties of such plasmas (4).

METHODS

Five microcuries of Few,3 as ferric chloride, were pre-pared for injection by introducing about 0.05 ml. of theradioactive iron solution (specific activity 1.3 mc./mg.)into a rubber capped test tube, to which enough 4% so-

dium citrate was added to bring the pH between 6.5 and7.0 (approximately 5 ml.). The mixture was autoclavedand 1 ml. of plasma fraction IV-7 containing 25 gm. ofprotein per 100 ml. was added with sterile precautionsand incubated with occasional shaking at 370 C. for 30minutes. During this period all of the iron combinedwith the protein, as proved by dialysis. Five ml. of thismixture were injected intravenously into suitable sub-jects in the post-absorptive state and the radioactivity of

2 Obtained from the Department of Physical Chemistry,Harvard Medical School, through the courtesy of Dr. E.J. Cohn.

SThe Fe" initially used in this study was obtainedthrough the courtesy of Dr. John H. Lawrence. The highspecific activity Fe" subsequently used in the cases ofmarrow aplasia was obtained on allocation from the OakRidge National Laboratories, through the Atomic EnergyCommission.

32

RATE OF REMOVALOF FE59 FROMTHE PLASMA

successive samples of plasma measured. There were nountoward reactions to this procedure.

Samples of plasma were prepared by centrifugation ofheparinized venous blood obtained at 10 to 30 minute in-tervals over a two to six hour period following the in-jection of the radioactive material. Total iron and latentbinding capacity of the plasma (5, 6) were determinedon at least two of the samples. Aliquots of 1 to 4 ml.were pipetted in duplicate into porcelain crucibles, to eachof which were added 5 mg. of carrier iron as ferric chlo-ride. Specimens were dried in an oven at 70° C. andashed in a muffle furnace at 600° C. for 10 hours. Theresidue was dissolved in a few ml. of concentrated hydro-chloride acid, evaporated almost to dryness and trans-ferred quantitatively with distilled water into electroplatingcells. A small amount of ascorbic acid was added to con-

vert the iron to the ferrous form; the solution was thenmade slightly alkaline with concentrated ammonium hy-droxide and buffered with 3 ml. of saturated sodium ci-trate. The iron was electroplated onto a thin copper disc2.5 cm. in diameter, using a rotating platinum anode, at1 ampere and 10 volts until the thiocyanate spot test foriron in the solution was negative (about four to fivehours).4 All the specimens were plated in duplicate.

The amount of radioactive iron electroplated on eachdisc was determined using an end window (mica, 1.5 mg./cm2.) Geiger Muller tube; the background was about1 count per second and the overall sensitivity was approxi-mately 5,000 counts per second per microcurie Fe".The counts per disc ranged from 3 to 100 counts per sec-ond and were determined by observing the time requiredto accumulate 8,192 counts.

RESULTS

The disappearance of the injected labelled fer-ric beta-l-globulinate from the circulating plasmawas exponential in all cases for the period of timefollowed. On semilogarithmic paper a straightline plot could be obtained and the rate of decreasein radioactivity determined by the slope of theline. For purposes of this report, this rate is ex-

pressed in terms of the half-time of disappearance(T%), i.e., the time in minutes for the radioac-tivity of the plasma to be decreased to 50%o ofits initial value.

Values for absolute plasma iron turnover per

unit time may be calculated from the radioiron dis-appearance data and the total plasma iron (3).Our results agree in general with those of Huffand his coworkers; several factors may influencethis absolute iron turnover and form the basis ofanother report.

Studies on eight normal adult individuals and32 patients with various blood dyscrasias comprise

4 This method was suggested by Dr. Paul Hahn.

this report. From the data obtained in the formergroup, the normal pattern of plasma iron disap-pearance was derived. No attempt was made toascertain the influence of various subsidiary fac-tors as age, weight, sex, metabolic rate, and -dietin essentially normal persons that could perhapshave some influence upon the plasma clearancedata. These factors may require elucidation inthe future. However, there is reason to believethat such variables will not have a decisive influ-ence upon the results in the present study.

The results obtained are summarized in TableI and Figures 1, 2, and 3. The half-time of disap-pearance of the radioiron in the normals rangedfrom 70 to 140 minutes with an average of 90minutes.

Three adult patients with chronic aplastic ane-mia (cases 9, 10, 11), in whom the diagnosis hadbeen made on the basis of the usual clinical and lab-oratory data, were studied. Each had received150 or more transfusions over a period of severalyears, and there was a uniform darkening of theskin and no splenomegaly. Skin biopsies revealedhemosiderosis. The serum iron was elevated andthere was complete saturation of the beta-l-globu-lin. The disappearance of radioiron from theplasma was markedly prolonged with half-timevalues of 220, 260, and 270 minutes, respectively.A fourth patient (case 12) with a hypoplastic ane-mia and a superimposed hemolytic component whohad received 15 transfusions totalling 7,500 ml. ofblood, similarly showed a prolonged half-time ofdisappearance of 215 minutes. A few weeks fol-lowing splenectomy, this value remained essentiallyunchanged at 225 minutes.

In addition, two patients, aged 10 and 22 years(cases 13 and 14) with a syndrome similar toaplastic anemia, namely, chronic aregenerative ane-mia, who had been transfused approximately bi-monthly since birth were studied. They bothshowed a paucity of red cell precursors in the bonemarrow, a severe normochromic anemia with nor-mal white cells and platelets in the peripheral bloodand the usual evidence of widespread hemosidero-sis. The serum iron was elevated in both and thebeta-l-globulin was completely saturated. Thehalf-time of disappearance in both these cases wasmarkedly prolonged to 205 and 245 minutes.

Another group of subjects showing bone mar-

33

L. R. WASSERMAN,I. A. RASHKOFF, D. LEAVITT, J. MAYER, AND S. PORT

row aplasia or hypoplasia but associated with mas-

sive splenomegaly and myeloid metaplasia of thespleen was investigated. This series consisted ofseven patients (cases 15 through 21) with myelo-fibrosis in whomthe diagnosis was proved by bonemarrow biopsy and splenic puncture. In six ofthese the rate of removal of radioiron from theplasma ranged from a half-time value of 20 min-utes to 120 minutes; i.e., 20, 25, 35, 95, 95, and120 minutes. Two of this group (cases 20 and 21)

received localized irradiation to the spleen (250rand 650r, respectively), following which therewas a decrease in splenic size; concomitantly, thetime of disappearance increased from 30 to 60and from 20 to 210 minutes, respectively. Thetwo patients with myelofibrosis (cases 18 and 15)who had received large numbers of blood trans-fusions, 125 and 110 units of blood, respectively,had skin biopsies which were positive for hemo-siderin; the serum irons were elevated and the

TABLE I

Age Hematocrit Transfusions Serum iron Plasma Fes'Case no. Diagnosis Sex ( ( (blood-mi. (gammaro%) disappearance

X105) .(Tj-ii.

NormalNormalNormalNormalNormalNormalNormalNormalAplastic anemiaAplastic anemiaAplastic anemiaHypoplastic anemia(post splenectomy)Aregenerative anemiaAregenerative anemiaMyelofibrosisMyelofibrosisMyelofibrosisMyelofibrosisMyelofibrosisMyelofibrosis(after splenic

irradiation)Myelofibrosis(after splenic

irradiation)Polycythemia veraPolycythemia veraPolycythemia veraPolycythemia veraPolycythemia vera(after 5 mc. P32)

Pernicious anemia(after vit. B12)Pernicious anemia(after vit. B12)Acquired hemolytic anemiaSickle cell anemiaChronic lymphatic leukemiaReticulum cell leukemiaChronic lymphatic leukemia(after cortisone)Hypochromic anemiaHypochromic anemia:MyxedemaUremiaLymphosarcoma (post HN2)Disseminated lupus erythematosusLymphosarcoma

* Beta-l-globulin 100% saturated.

1

2345

6789

1011121213

1415161718

'192020

2121

22232425262627272828293031

32333334353637383940

M

M

M

M

M

M

M

M

M

FFM

M

M

M

M

FFFFM

M

M

M

M

M

M

M

FFFFFFFFFFFFFFM

FFFM

44464949383250

521660524949102239425662467272-J

5353

3050374751517676747451484644606030554959604053

484050

42414346403013323038312624282123363232

3127

5280635351483137263822222621353727354024272221

0

0

0

0

0

0

0

0

75

115

105

7.5

7.5

160

140

110

0

1.5

125

0

0

0

2.5

2.5

0

0

0

0

0

0

1

1

0

0

5

5.5

1.5

35

7.5

8.0

0

0

0

0

2

0

2.5

5090

11095

13011080

100153*205*288*272*212*268*178*185*84

137193*

8511290

125143

105113

55120120130235*118265*

78150188108118102112

2625

148195115

77133

100708070

1401009575

2202602702152252052452131209595352560

20210

6020114530

145409030552535

120265

68.36306085

185290

8885

I___ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ I I____ I___ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

34

RATE OF REMOVALOF FE5 FROMTHE PLASMA

L20

iaW

0LU!;.

LJLAh

0

A

20 40 6080 100 12o 140 160 180 20 220 so260 0

TME(MIN. AFTERLV Fe"-BETA-I-GLO WLINATE)FIG. 1. COMPARISONOF THE PLASMARADIOIRON DISAPPEARANCEIN THREE

PATIENTS WITH APLASTIC ANEMIA (CASES 9, 10, 11), T%= 220, 260, AND270MIN', WITH THREEPATIENTS WITH POLYCYTHEMIAVERA (CASES 24, 23, 26),T3h = 11, 20, AND 30 MIN.

Shaded area is normal range with average T/ = 90 min. Note the very rapidhalf-times of disappearance in the presence of hyperactive erythropoiesis, andthe markedly prolonged TMwhen red cell production is diminished or absent.

latent binding capacities of the sera were zero.

In the former case, the half-time of disappearancewas 95 minutes while in the latter it was markedlyprolonged to 210 minutes. This patient was theonly member of the myelofibrosis group to havean abnormally prolonged T%. Subsequent post-mortem examination revealed marked myeloidmetaplasia of the spleen and widespread general-ized hemosiderosis.

In two cases of chronic lymphatic leukemia thehalf-times of disappearance were 120 and 68 min-utes. Following one month of cortisone therapythe value in the latter subject (case 33) had de-creased to 36 minutes. During this time there was

a reticulocytosis of 9%, a slight increase in thepacked cell volume and in normoblasts in the bonemarrow. Another patient with reticulum cellleukemia who had been splenectomized and hadreceived 70 transfusions, had a T% increased to265 minutes; a sternal marrow examination doneat that time revealed only 3% of erythroid ele-ments.

The clearance of the tagged iron from the plasmawas increased in polycythemia vera. Five pa-

tients with the typical clinical and laboratory find-ings of idiopathic polycythemia had half-timevalues of 11, 20, 30, 45, and 60 minutes. One ofthese patients (case 26) was treated with 5 milli-curies of radioactive phosphorus and one monthafter the injection, the T%became increased from30 to 145 minutes.

A patient with acquired hemolytic anemia (case29) who had a severe anemia, splenomegaly, re-ticulocytosis, increased urinary and fecal urobilino-gen, a positive Coombs' test and a markedlyerythro-normoblastic marrow, had a T% of 25minutes. In another patient with the hemolyticsyndrome of sickle cell anemia, the T% was 35minutes.

Two women with severe hypochromic micro-cytic anemia (cases 34 and 35) and low serum ironconcentrations had rapid half-times of disappear-ance of 30 and 60 minutes.

Two patients over 70 years of age with per-nicious anemia in relapse who had severe macro-cytic anemia and combined system disease, showedincreased half-time rates of 30 and 40 minutes. Sixweeks following continuous therapy with vitamin

35

36 L. R. WASSERMAN,I. A. RASHKOFF, D. LEAVITT, J. MAYER, AND S. PORT

92 POTYC9THEMIAVERA

o ~~~~MLICRE\J ITAVTER.US32

0~~~~~~

,U2 \B, FOI. P32. . ,T IP3*

-LL20 40 60 80 W 120 940 960 980Z~ ?2 240 26020AL

TIMlE(MI N.AFTER. LV. Fe5 BETA+G4LOWULINATE)FIG. 2. COPATIENTWIT POLYCTHEVAF-IERAF (CASAPEA26RWHNRCEIE5FPAM

MILLICUBiESPMRI0NTMAV U

NoeteicesFnteT/rom a pretreatmenthaftm value of 30minm.lwig60 to 145

mp.eon, te montheaanclae This sraots te non10efec of 0idporessng

Notethe inreenase insthefor frmartrelat enyth value of 30thin.tydom 145mhsindicatedmonth lterhis influstraeso the knowneff ofred clprdtion deprssin

erythropoiesis.

I00%9 ~~~~~~MYELOFIBPOSISZ 6 .*AFTER SPLENIC IRRADIATION

Q~~~.4~ ~LU

Tk 9O"C '

Lii 2 \BEFORESPLENIC1= ~~~IRRADIATION

20 40 608 100 120 14 16 £50 MM2OzMO

TIME (MIN. AFTER. V Fe" BETA+1-GLOBULINATE)FIG. 3. COMPARISONOF THE HALF-TIME OF DISAPPEARANCEOF PLASMA

RADiorRON BEFORE AND AFTER SPLENIC IRRADIATION IN A PATIENT WITHMYELOFIBROSIS ANDMARKEDSPLENOMEGALY

From a pretreatment half-time value of 20 min. following 650r to thespleen, the disappearance rate became prolonged to 210 min. The importanceof the spleen as a site for extrai-medullary erythropoiesis in this syndrome isthus indicated and the influence of the degree of red cell production on theplasma radioiron clearance is demonstrated.

RATE OF REMOVALOF FE59 FROMTHE PLASMA

B12, the anemia improved, the marrow becamenormal and the T%became prolonged to 55 and90 minutes respectively.

A miscellaneous group of patients was also ex-aimed. The half-time of disappearance in thevarious cases was as follows: severe myxedema,85 minutes; uremia, 185 minutes; disseminated lu-pus erythematosus, 88 minutes; lymphosarcoma,85 minutes. Another patient with lymphosar-coma who had received 2,000 ml. of blood, was

given nitrogen mustard therapy; two weeks laterthe T% was greatly prolonged to 290 minutes.A sternal marrow examination done before treat-ment was normally cellular with 20%o erythroidelements. Following HN2 therapy at the time ofthe studies with radioiron, the marrow was mark-edly hypoplastic with only 3%o of red cell pre-

cursors.

DISCUSSION

In early experiments, the radioiron was incu-

bated in titro with the patient's own plasma priorto its intravenous injection. Although this tech-nique gave reliable results in many individuals, itwas observed that in subjects with hemosiderosis,the radioactive iron was, lost from the plasma ata-very rapid rate. This was found to be due to thefact that the beta-l-globulin of the plasma of thesepatients was already saturated with iron, hencecombination between the radioiron and the plasmaglobulin did not occur. The iron injected thusremained in the ionic form and its rate of disap-pearance from the blood stream was markedly dif-ferent from that of iron bound to globulin (4).The initial rapid loss of the tagged iron is prob-

ably due to diffusion into the intercellular space

whereas the subsequent exponential disappearanceof globulin bound iron results in part from the in-corporation of the metal into hematopoietic foci,preparatory to hemoglobin synthesis, and deposi-tion into the miscible iron pool.

To obviate this variation in combining capacityof different plasmas, a fairly pure source of beta-i-globulin was utilized to prepare the iron proteinate.The plasma fraction IV-7 (Cohn) as supplied isconcentrated approximately 70-fold over normalplasma with respect to the metal combining globu-lin. When the radioiron is incubated with frac-tion IV-7 all radioiron disappearance rates are ex-

ponential irrespective of plasma iron concentra-

tion and the concentration and saturation of thecirculating beta-1-globulin.

The nature of the radioiron disappearance curvedeserves comment. Although no data for any pe-riod greater than six hours are available due toexperimental limitations, it seems unlikely thatany deviation from the exponential removal wouldoccur. If this should happen, one would have toassume either an interruption of the steady stateof the iron cycle or a feed-back of injected ironwhich has left the plasma. In the former instance,a new equilibrium would probably be reached andthe function would again become exponential al-though possibly with a different rate constant.If feed-back is considered a possibility, this wouldimply a completion of the marrow or tissue storecycle which is highly improbable for the first fewweeks, at least.

The rate oot disappearance may be an average ofseveral similar rates which are in turn dependentupon the uptake of iron by separate end organs,i.e., the erythropoietic tissue, the fixed iron storesand the available iron pool; or perhaps in a particu-lar metabolic state, one rate, exponential in nature,so dominates the plasma iron turnover that otherrates become insignificant. Thus, in marrowaplasia where only a few percent of an injecteddose of radioactive iron ultimately appears in thered cells, the disappearance rate is influencedlargely by the uptake of iron by the tissues; ironleaves the plasma at a very slow rate in the blooddyscrasias in which the production of red cells isgrossly impaired, as in aplastic and aregenerativeanemia (7-9). On the other hand, in marrow hy-perplasia, most of the radioiron becomes incor-porated into the circulating erythrocytes (7-9);the rate of disappearance in this case is dependentin the main upon erythropoiesis, so that the pres-ence- of greater than normal numbers of red cellprecursors, such as are known to exist in the bonemarrow of polycythemia vera, hemolytic anemiaand iron deficiency anemia, is accompanied by arapid clearance of iron. These findings suggestthat the turnover of the plasma iron as measuredby the radioactive technique is an index of the de-gree of erythropoiesis in the body.

In pernicious anemia in relapse, in view of themegaloblastic arrest of the marrow a decreasedrate of plasma iron disappearance might be antici-

37

L. R. WASSERMAN,I. A. RASHKOFF, D. LEAVITT, J. MAYER, AND S. PORT

pated. However, the values in two patients (cases27 and 28) were quite rapid, 40 and 30 minutes,respectively. Six weeks following vitamin B12therapy when the bone marrows were normal andthe peripheral blood had returned nearly to normal,the study was repeated and respective rates of 90and 55 minutes were obtained. In view of the highserum iron, complete saturation of the metal com-bining globulin and tissue hemosiderosis that existsin this disease, abnormal and rapid iron storagemay play a role in the increased rate of plasma irondisappearance. Another possible explanation isthe avidity of the marrow for iron in these casesbecause of actual increased erythropoiesis. Thegross morphologic picture of red marrow replacingthe normally yellow marrow areas coincides withthis idea. It is possible that the erythrocytes andred cell precursors which are formed in perniciousanemia are broken down in situ. This is suggestedby an excretion of urobilinogen far above thatwhich would be derived from ordinary hemolysisin this disease. In this regard, London andWest (10) have shown, using N15 labelled glycinein a case of pernicious anemia in relapse, that 40%of the urobilinogen appearing in the stool was notderived from circulating erythrocytes.

The bone marrow in myelofibrosis is hypocel-lular and a large portion of the marrow cavity isreplaced by fibrous tissue or bone. There is usu-ally a progressive splenomegaly with the spleengradually assuming the erythropoietic function.Because of this compensatory metaplasia in thespleen, relatively good hemoglobin levels may bemaintained for long periods of time. In view ofthe hypoplastic, fibrotic bone marrow, a slow rateof clearance of iron could be expected; however,most of the rates were rapid or within the normalrange in the cases studied. This may be due toactive red cell production in the spleen with, insome cases, a concomitant increased destructionof erythrocytes to account for the anemia. An ele-vation in the fecal and urinary urobilinogen waspresent in these latter cases and is suggestive con-firmatory evidence of such a process. One of thesepatients (case 21) who had received a total of650r to the spleen in a three week period, diedsubsequently. At autopsy the spleen showed veryslight erythropoiesis contrary to that usually ob-served in myelofibrosis of advanced degree withmyeloid metaplasia in the spleen. The prolonga-

tion of the half-time of disappearance of theplasma radioiron by splenic irradiation in the twopatients mentioned above, demonstrates the effectof erythropoietic activity on plasma radioironclearance. Furthermore, it emphasizes the im-portance of the spleen as a blood forming organ inthis syndrome. The only patient of this groupwith a prolonged rate of disappearance was case15, in whomthere was a marked degree of exoge-nous hemochromatosis, secondary to multipleblood transfusions.

The influence of increased iron stores upon therate of radioiron disappearance from the plasmais important. Data bearing on a closely alliedproblem have been reported by Finch and associates(7, 11). They found that in patients with en-dogenous hemochromatosis, radioiron utilization isprofoundly depressed while erythrocyte produc-tion proceeds normally. Similarly, in normal sub-jects fed large quantities of iron for fairly longperiods, the utilization of iron is reduced. Again,in iron loading experiments in dogs, comparisonof pre- and post-loading utilization curves showeda markedly decreased uptake in the latter. It ap-pears fairly definite therefore that increased ironstores depress the utilization of radioiron for he-moglobin production. Increased iron stores maysimilarly affect the rate of disappearance of radio-active iron from the plasma. Our data are as yettoo meager to resolve this question. As has beennoted above, case 15 with myelofibrosis and hemo-siderosis exhibited a markedly prolonged T%;however, another subject (case 18) with the samecombined syndrome, who had received an evenlarger number of blood transfusions, had a normaliron disappearance rate. All the patients witharegenerative and aplastic anemia had hemosidero-sis, but it appears more likely that the marrowaplasia rather than the enlarged iron depots is theprimary factor in prolonging the disappearance ratein these cases.

Although it is possible that iron stores, may af-fect the radioiron disappearance rate, it has beendemonstrated that marked variations in plasmairon clearance may occur in the same patient withno change in iron stores. Thus, the patient withpolycythemia vera (case 26) who received 5 milli-curies of P82 exhibited a pretreatment of T% of30 minutes which then became prolonged to 145minutes one month later. At the same time, the

38

RATE OF REMOVALOF FE5 FROMTHE PLASMA

packed cell volume decreased from 51%o to 48%band the per cent of erythroid elements in the sternalmarrow similarly showed a reduction from 60%oto 25%o. The decrease in the half-time of disap-pearance was then a gauge of the effectiveness ofbone marrow irradiation in depressing erythro-poiesis. In view of the very slight decrease in he-matocrit and no essential change in serum ironconcentration (see Table I), it is clear that no al-teration in iron stores was likely.

As has been previously noted, two patients withmyelofibrosis had received localized irradiation tothe spleen immediately following which the radio-iron disappearance rates decreased. No bleedingwas observed nor were any transfusions admin-istered in the interim. During this period, thepacked cell volumes decreased moderately whilethe serum iron remained essentially unchanged.Here again in the presence of constant iron -storesa marked decrease occurred in the turnover of ra-dioactive iron from the plasma.

It is apparent that the removal of radioiron fromthe plasma is an index of red cell production inthe body inpatients with normal iron stores. Achange in the steady state of the turnover of ironwill produce an effect on the iron clearance fromthe plasma. Following the establishment of a newequilibrium, a new iron disappearance rate con-stant will be obtained. The rate of iron disap-pearance may be of value as a diagnostic test, par-ticularly when combined with a study of the utiliza-tion of iron by the red cells. However, wherelarge numbers of transfusions have been given, therate of disappearance may be influenced by the in-creased fixed iron stores. The latter problem iscurrently being investigated in this laboratory.

SUMMARY

1. A method for the determination of the rateof disappearance of iron from the plasma usingradioactive iron and a solution of beta-l-globulinhas been described.

2. A correlation between erythropoiesis and therate of disappearance of radioiron from the plasmahas been established.

3. In aplasia of the bone marrow, the rate ofremoval of radioiron from the plasma is markedlyprolonged. Increased marrow erythropoiesis asoccurs in polycythemia vera, hemolytic anemia

and severe hypochromic anemia results in anincreased rate of removal of radioiron from theplasma.

4. Greater than normal iron stores, as occursin hemosiderosis, may prolong the plasma radio-iron turnover.

5. It is suggested that with normal body ironstores the half-time of plasma radioiron disappear-ance is a sensitive measure of the integrity of theerythropoietic tissue of the body.

REFERENCES1. Flexner, L. B., Vosburgh, G. J., and Cowie, D. B.,

Capillary permeability; rate of transcapillary ex-change of iron added to plasma as radioactive ferricbeta-l-globulinate. Am. J, Physiol., 1948, 153, 503.

2. Huff, R., Hennessy, T., and Lawrence, J. H., Ironmetabolism studies in normal subjects and in pa-tients having blood dyscrasias. J. Clin. Invest.,1949, 28, 790.

3. Huff, R. L., Hennessy, T. G., Austin, R. E., Garcia,J. F., Roberts, B. M., and Lawrence, J. H., Plasmaand red cell iron turnover in normal subjects andin patients having various hematopoietic disorders.J. Gin. Invest., 1950, 29, 1041.

4. Wasserman, L. R., and Rashkoff, I. A., Unpublisheddata.

5. Rath, C. E., and Finch, C. A., Chemical, clinical, andimmunological studies on the products of humanplasma fractionation. XXXVIII. Serum irontransport. Measurement of iron binding capacityof serum in men. J. Clin. Invest., 1949, 28, 79.

6. Cartwright, G. E., and Wintrobe, M. M., Chemical,clinical and immunological studies on the products ofhuman plasma fractionation. XXXIX. The ane-mia of infection. Studies on the iron binding ca-pacity of serum. J. Clin. Invest., 1949, 28, 86.

7. Finch, C. A., Gibson, J. G., II, Peacock, W. C., andFluharty, R. G., Iron metabolism. Utilization ofintravenous radioactive iron. Blood, 1949, 4, 905.

8. Dubach, R., Collender, S. T. E., and Moore,. C. V.,Studies in iron transportation and metabolism.VI. Absorption of radioactive iron in patients withfever and with anemias of varied etiology. Blood,1948, 3, 526.

9. Rashkoff, I. A., Wasserman, L. R., Leavitt, D;, andMayer, J., Radioactive iron studies in anemia. Bull.N. Y. Acad. Med., 1951, 27, 387.

10. London, I. M., and West, R., The formation of bilepigment in pernicious anemia. J. Biol. Chem.,1950, 184, 359.

11. Finch, C. A., Hegstead, M., Kinney, T. D., Thomas,E. D., Rath, C. E., Haskins, D., Finch, S. and Flu-harty, R. G., Iron metabolism.. The pathophysiologyof iron storage. Blood, 1950, 5, 983.

39