Embed Size (px)
Citation preview
THE LOCAL BLOODFLOWIN HUMANBONEMARROWINLEUKEMIA ANDNEOPLASTICDISEASES AS DETER-
MINED BY THE CLEARANCERATE OFRADIOIODIDE (I181) 1
By N. L. PETRAKIS, S. P. MASOUREDIS,2 AND PATRICIA MILLER, WITH THETECHNICAL ASSISTANCE OF FAUNOL. CORDES
(From the Laboratory of Experimental Oncology, National Cancer Institute, National Insti-tutes of Health, Public Health Sertice, Department of Health, Education, and Welfare.
and the Division of Medicine and Cancer Research Institute, University ofCalifornia School of Medicine, San Francisco, Calif.)
(Submitted for publication December 1, 1952; accepted June 6, 1953)
Although considerable information has beenobtained on the rate of blood flow in many or-gans (1), few data have been reported on theblood flow in human or animal bone marrow (2).The neglect of this aspect of marrow physiologyis probably due to the technical difficulties en-countered in gaining access to the blood vessels ofthis diffuse organ in the intact animal or man.
The extensive application of the radioisotopeclearance technique described by Kety (3, 4) forthe estimation of the regional circulation inmuscle, liver, subcutaneous tissues, brain tissue,and skin (4-10) suggested its use as a clinicallyapplicable method for the evaluation of the effec-tive local blood flow in the human bone marrow.This method is based on the premise that if adiffusible tracer substance is introduced into atissue its clearance or removal is a function of allpossible factors contributing to the local bloodflow at the site of injection.
In the present study, the relative rates of clear-ance of II8' from the bone marrow were deter-mined in patients with various malignant dis-eases and leukemia. The radioiodide was in-jected into the bone marrow through a standardbone marrow needle, and the rate of disappear-ance of the isotope was measured with a Geiger-Mueller counter. In addition, simultaneous de-terminations of I'll clearance rates from the gas-trocnemius muscle were obtained for comparingthe clearance rates obtained in the bone marrow
1 Presented in part at the Annual Meeting of WesternSection of the American Federation for Clinical Re-search, Carmel, California, January 24, 1952.
2Postgraduate Research Fellow, National Cancer In-stitute.
with those found in an anatomically and physio-logically different vascular bed.
MATERIALS AND METHODS
Clinical material: Forty-three patients with a varietyof malignant diseases and leukemia were studied (TableI). The patients were grouped into seven major groupsas indicated in Table I. They varied in weight from28 to 91.6 kilo, and all had normal blood pressures ex-cept M. D. and E. H., whose pressures were 175/110and 160/110, respectively (Table I). The subjects werestudied under resting but not basal conditions. Theenvironmental temperature varied from 20 to 250 C.
Marrow technique: The sternal marrow and thegastrocnemius muscle were studied simultaneously ineach patient (Table I). In a few subjects the tibia andthe ilium were employed. The iliac crest site was ap-proximately two centimeters posterior to the anteriorsuperior spine of the ilium and the tibia was studied atthe tibial tubercle.
The marrow site was aseptically prepared and locallyanesthetized with 2 per cent procaine infiltration of theskin, subcutaneous tissue and periosteum. A shortTurkel needle was advanced into the bone with a rotarymotion until the marrow cavity was entered. The radio-iodide was then introduced into the marrow tissue bymeans of a tuberculin syringe with a No. 22 gage 4-inchneedle inserted through and embedded well beyond thetip of the bone marrow needle. The I', as sodiumiodide, was slowly injected4usually requiring one to twominutes. A slow rate of injection was used to minimiizepain and to avoid a sudden increase of intramedullarypressure which conceivably might affect the rate ofclearance of the I`m By placing the point of the No. 22gage needle well into the marrow and slowly injectingthe radioactive material, it was possible to prevent skincontamination by reflux of marrow blood and I' intothe marrow needle. Following the injection of I' bothneedles were withdrawn from the marrow and pres-sure was immeditely applied to prevent bleeding into thesubcutaneous tissues and onto the skin. The puncturesite was cleaned with alcohol to remove possible radio-
952
BLOOD FLOW IN HUMANBONE MARROW
'000- t--'Inooo rovC
_0_0'0 0000%000% '00%0t
080to 0000
000b>> 00%r
bo 0 U)D 0
ooco U) Oi) 00t.cj00t0 00 .
+4 00Vt 00 (N u000 4 C%tV
0
004-0% -0if) 0o0 '0 '-u -z ~ -_ -rcl 6~-:
* * * *
0 t' (N 000 (N - (N 00odo6ooooo
8800000008.
N N
_
N 'CA 0 000 00--i - ------66 l;.
'-6000 0%0%4'-0'o0(4to00 evio4400'
4t)
b0o.XV24
%p3al
bi 9: M F!j
E:40:~t -~ d
4-'42
040e
00 C4 'O- - 0
C o6 oo~'N In- In tll*
- 00-+Cc'q C00 t
o O 00
004O4%*6 4c-
WI) .40 " " - U) oo r
14 -
,i C4( 00 4 - 00-00
8 8000000 000000 '- 0-0 0 "-4-
W-
V- 0 OlJ 00%qmif
-.. -i -~l4
(N -it t-00o (N- (N'IC 0c4No
(. ,~0 ijO I
~414
4)
.4)4-'
80XC)040
U4L
o N '0el; 00o o o00 '0- 0 - r Nt- -000-00-0
4-1.00U)_( 00 'N00U '0CD I0'0t-00(N4--0 0N60adi1- 6o c 6~ 0C;
0 00
1;04- 6 C-i u-4 40-' e4
" W. W) V- V-- V- -4M00000000* * .00000000oo 00 (
4-t- N'00% 0(N4 4
.Rd 00 4- '0 (N. 4- '00 ON - 4'. 00
.4uas
o6 a-Z 4 o Z o6o6re;- - (N - --.
00 0t
i00 %40 0ael; el;
o o0lvW)9 l)u Ul)no"M Mc+nol)*00040 000%0000o
000000008o000o000ooo4'-000004'-O0000n100o0 -0
ot'-
0(N4'-00o
0% -4 - 00'- '0 ' (N 00
el-~C r _ l--- -Z 6
000000OS 4- 0% 000000'0
4-b 4. JJ4.b 4.64- e41 4
c.4 :g
&:9: :11 4;,
953
WW09 v
.t
El
IOU
I
'3
II
P4
10.4
.J!
a
IVI.
eC.
-t-II
rt.%b
aQ
.9-3n
" r.94-1m I(.1 1
43P.&
.8Vb
:9t4
1%&
v
I
-Q44
lbql-.E.%IeA
4)I
4)*0
z
I 01- 1-
L) gm.: I
9.0%m14
ll-.-t
i
N. L. PETRAKIS, S. P. MASOUREDIS, AND PATRICIA MILLER
0C CUI) 00 00 00 0%
6+ono~ o4
* *.*." - *..
(Ito 141 0 00b io
C4N0 m4 -4m)t t I
oo" oC4
- e
(N
0011)000 (U) 000%~ k-- :
~o CN C4 Zb e
(U)11)11)0 0 00
00000 00
(Nj 00 4U-0j4-
***** *W
oooo0% oo0~
ooooo0 oo(
oooooU oo.4.1.1414.4. 4.4.
,8 88U)U)U)U)U U)U
*OQ(*DQ OC.
00000 00O
QUOQ..O U(O.
110 i~ U,ig
954
Ch
ne0
4-1-Vo4 ONU)ui
0 011) (N
00(U -
66
4-00 00 lq4 'O0000%
* *
6 (Ii tf;
10 I-) 00 00
C)(N (Nj CN
*. C. 0 0.
00 00
0nooUi
-o +4-4(N
(U 411 U
U)M.XC :
0Y(U2U) .U*U' $E-
U)
0
0(U
ICH.
U)
oC
*g0
CU
C._
3
0
._z
CL)._
4-
U)
UL)
C-)
80.0o
la
CUU)
U)o
0
0.4-
I;
t- It) ' 00
l o
011)0t 00
Coo2* * . .
0sncsb EoO
.~r, -I-C'ouo oo1
0U)0o666 00
*00 '0 (U) 0
in) 'U ) 1)0 11
11)
(Nj 0% -4(N (4
~o ul o Nt .Ct
000000
11 U) If 00^o
-(U)Ov
*000.%.0. -
0% 0% 11)0 U-CU
; . * **~ .3 . .: P@
~-. c0.
0 U)
*) 4-44I
.;Cj I:i . 04. w<..; 1:41:4
BLOODFLOW IN HUMANBONE MARROW
active contamination, and the wound was sealed withcollodion.
Marrow was obtained for cytological examination atthe end of counting procedure in most instances. In afew patients, however, 0.1 to 0.2 cc. of marrow was ob-tained at the start of the study without demonstrablealteration in clearance. In the later instances care wastaken to inject the I' well beyond the point of the as-piration. Routine differential counts of the marrowsamples were made with both supravital and Wrightstained fixed preparations. Usually 200 cells werecounted for the differentials.
In most patients, I clearance rates of the gastrocne-mius muscle were simultaneously determined, exceptwhere noted (Table I). The radioiodide was injectedinto the belly of the muscle by means of a tuberculinsyringe with a No. 26 gage llh inch hypodermic needle.
Physiologic and pharmacologic studies designed todemonstrate the effect on I' clearance of altered sys-temic circulatory dynamics were carried out in a fewsubjects. These included exercise, phlebotomy and theintravenous injection of Paredrine 3 and of Bistriumbromide (Squibb).4
Determination of radioactivity: From 100 to 400 micro-grams of iodide and from 10 to 30 microcuries of I'were used in all determinations. The radioiodide wasmade up in isotonic phosphate buffer, pH 7.5, solutioncontaining carrier potassium iodide. The volume ofradioiodide injected per patient is shown in Table I.Surface counting was carried out with a lead-shielded(1.25 cm. thick) Victoreen or Tracerlab mica end-window counting tube mounted on an adjustable stand.The window thickness of the tubes varied from 3.5 to1.8 mg. per cm.'. The window-to-tissue distance was0.5 cm. In addition, two layers of gauze were interposedbetween the counting tube and the skin over the marrowsite. The radiation detected in the marrow under theseconditions of geometry is predominantly gamma radia-tion. The Geiger-Mueller tubes were coupled to twoconventional scalers, one tube being used for the marrowstudy and the other for the muscle determination. Eachinjection site was counted for a total of 1,024 counts,every 30 to 60 seconds. The counting was carried outfor fifteen to thirty minutes after injection. In all casesthe counts were greater than ten times background.
Failure to obtain a decrease in counting rate withrespect to time was indicative of gross skin contamina-tion by radioiodide. The high efficiency of our tube forbeta radiation and the absence of a physiological mech-anism for removal of I' from the skin provided anextremely sensitive measure of skin contamination.Where skin contamination occurred, an attempt wasmade to resume counting after careful cleansing with 70per cent alcohol. Data were rejected if counting couldnot be started within three minutes after the injectionof I. Data were also rejected if a single exponentialdecrease did not occur within the first 15 to 30 minutes
8 B-4-hydroxyphenylisopropylamine.4 Hexamethonium bromide.
after inj ection. Skin contamination not sufficient toobscure marrow clearance was manifested by failure toobtain a single exponential decrease, since the I' on theskin would contribute an essentially constant number ofcounts to the counting rate during the time of the study.Skin contamination did not occur with the muscle studiessince the I' was injected into the muscle from a lateralposition. This enabled counting to be done over skinthat was out of the puncture site.
Studies were made to ascertain whether the radioac-tivity being measured was actually in the bone marrowand that the clearance curves were not significantly in-fluenced by extravasation of minute amounts of bloodcontaining radioiodide into the subcutaneous tissues.Marrow clearances were made employing aluminum-shielded (1.96 mg. per cm.') Geiger tubes. Countingwas continued until the clearance curve was established,after which the aluminum shield was removed and thecurve again determined. The clearance curves ob-tained in this manner were not significantly different,indicating that the radioiodide was injected into the bonemarrow and that practically no contamination of thesubcutaneous tissue occurs under the conditions of thetechnique employed here. The error that results fromsuch extravasation would decrease the clearance ratebut would not detract from the positive findings re-ported here.
Calculations: The activity in counts per second wasplotted against time on semilogarithmic paper. In allcases studied, the results could be described as a single,decreasing exponential function. While other studies(8) have indicated that the disappearance curve is acomplex function, our studies were based on the initialcomponent which was logarithmic and extended throughthe first 15 to 30 minutes. In a few cases other com-ponents developed later. The present studies were com-pleted within 20 minutes in most cases. The results areexpressed both as half-time in minutes (T%), and asclearance constant (K), which is the per cent ad-ministered radioiodide removed per minute, as obtained
from the expression K= 0.63
To determine the statistical significance of the clear-ance constant derived from the plot of the data onsemilogarithmic paper, the regression line was deter-mined and the standard deviation of the half-time of theregression line was calculated (Mather, K, Statisticalanalysis in biology, Chapter VIII. Interscience Publ.Inc., New York, 1947) in four representative patientsfor marrow and muscle. It can be seen that the ex-perimental data fit an exponential function and thatthere is a significant difference between marrow clear-ances of different patients as well as significant dif-ferences between the marrow and muscle clearances insome of the patients (Figure 1).
Radiation dosage calculations: For calculation of theradiation received by the bone marrow, certain simpify,-ing assumptions were made:
955
N. L. PETRAKIS, S. P. MASOUREDIS, AND PATRICIA MILLER
IODIDE (i"'?) CLEARANCEFROM5TERNAL MARROWAND GASTROCNEMIU5MUSCLE
Chronic Lymphocytic L.uk.mna>^ M.P, Age 65, F
Acute MonoblaXtic LeulnsAC., Ag. 44, M.
* Marow -
Th*-.5Wt0.60i
Muscle-,TYz2z2.92O.46mur
10 15,5 0 5
TIME, IN MINUTESFIG. 1. COUNTSPER MINUTE OF Il PLOTTED SEMILOGARITHMICALLY
AGAINST TIME FOR STERNAL MARROWSAND GASTROCNEMIUSMUSCLESIN FOURPATIENTS WITH LEUKEMIA
The half time of the regression line and the standard deviation of thehalf time are demonstrated.
(a) The depth of the sternal marrow cavity was
assumed to be approximately one centimeter.(b) Since less than 1 ml. of a solution containing less
than 25 microcuries of I' was administered, it was as-sumed that the greatest amount of radiation was con-
fined to 1 cm.' of bone marrow tissue.(c) In this volume the gamma ray energy contribution
was considered to be negligible.
(d) No selective metabolic localization of I' oc-
curred in the bone marrow.
(e) The large dilution of I' in the systemic iodidespace, and the relatively rapid excretion of I' (11),contributed only a small amount of the radioactivity tothe marrow by recirculation and was considered insig-nificant.
(f) The effect of physical decay of I' was considered
K
z1~
956
BLOODFLOW IN HUMANBONE MARROW
insignificant (half-life eight days) because of the muchmore rapid rate of removal of the I' from the marrow
site.With these assumptions the irradiation to 1 gram of
bone marrow from beta-particles with 30 microcuries ofI' is 0.260 rep. per min. (10). The total accumulateddose (complete removal of isotope from the site ofinjection) is 3.70 rep. Only about 0.07 per cent of thebone marrow receives this radiation, assuming a marrow
volume of 1,400 cc. Clinically no evidence of hemato-poietic dysfunction attributable to the radiation used inthese studies could be detected.
The thyroid uptake of I' resulting from marrow andmuscle studies was determined in four patients at 6 and24 hours, and was found to range from 15 to 25 per centof the total administered dose. This represents 6 to 10microcuries.
RESULTS
Forty-three patients with various neoplasticdiseases, lymphoblastomas, and leukemia were
studied. The half time (T%) and K values (percent administered radioactivity removed per min-ute) for the bone marrow and muscle in eachpatient are presented in Tables I and II. Repre-senting curves from four patients are shown inFigure 1. The distribution of the individual Kvalues in the leukemic and non-leukemic patientsis shown in Figure 2. In a patient with Hodg-kin's disease (E. C.), simultaneous clearancestudies on the sternum and iliac crest yielded al-most identical results. The K values were
closely grouped in the non-leukemic patients andin the patients with lymphoblastoma, with an
average of 7.3 per cent IlJs cleared per minute,ranging from 2.86 to 12.6 per cent. In the acuteleukemias the average K value was 16.9 per centper minute, ranging from 12.6 to 20.4 per cent,indicating that the injected iodide was removed
from the marrow at a more rapid rate than innon-leukemic patients. In chronic lymphocyticleukemic patients there was a spread of K valuesfrom 7.29 per cent to 21.7 per cent per minute.Most of the K values for the myelocytic leukemiapatients were in the range of the values obtainedin the non-leukemic patients, 4.78 to 8.66 per centper minute, except for patients J. B. and A. J.who had more rapid clearance rates. The twopatients R. J. and J. M. with monocytic leukemiahad clearance values of 9.63 per cent and 5.10per cent, respectively. In four patients withmultiple myeloma, three (N. P., G. V., and V.D.) had elevated K values, 15.4, 16.3, and 15.0per cent, respectively, while the fourth patient(G. C.) had a low rate of 3.96 per cent.
There was no apparent correlation between themarrow clearance rates and the hemoglobin level,leucocyte count, weight, body temperature, andblood pressure (Figure 2). Although the vol-ume of J131 in these experiments varied from0.05 to 1.0 ml., there was no clearly demonstrablerelationship between volume of injected 1"81 andclearance rate. Rapid clearance rates were ob-tained with both large and small volumes of I"'.The average volumes injected are indicated inTable I, and it can be seen that they did not vary
more than 0.3 ml. for the marrow studies, and0.2 ml. for the muscle studies.
In all the patients studied, the muscle K valueswere similar, approximately 8 to 10 per cent ofthe JIS" removed per minute (Tables I and II).In the acute leukemia group, muscle clearanceswere obtained on only two patients, 4.62 and 5.37per cent, respectively. One patient with mono-
cytic leukemia had a muscle clearance of 18.2 per
cent Il-i removed per minute.
TABLE II
Averages of marrow and muscle clearance constants and ranges in 43 patients
Bone marrow Muscle
No. Average No. Averagecases K Range cases K Range
Non-leukemic 9 7.37 4.33-12.6 6 7.89 5.46- 9.90Lymphoblastoma 7 16.6 2.86-10.2 4 9.72 8.15-10.314.06 ~~~~~~~~~~17.61
Av. 7.03 Av. 8.31Lymphatic leukemia 10 13.41 7.29-21.7 9 9.16 6.03-19.3Myelocytic leukemia 7 10.3 4.78-20.4 9 9.65 6.08-17.3Monocytic leukemia 2 7.36 5.10- 9.63 2 13.7 9.36-18.2Multiple myeloma 4 13.4 3.96-16.3 3 9.77 7.97-12.1Acute leukemia 5 16.9 12.6 -20.4 2 4.16 4.62- 4.70
957
N. L. PETRAKIS, S. P. MASOUREDIS, AND PATRICIA MILLER
NOW-LIUKEMIC MAUGMNACIES24
20-
'5-
010_ * 0. ~ ~~~*s If
5-
0
S.
* .
ACUTE LEUKE"A - 0MULTIPLE MYELOMA-O
00000,
0
10 15
CDIONiC LYMHOCYrIC LEUKEMIA
0
0
00
CHRONIC MYELOCYTICLEUKEMIA-MONOCYTIC LEUKEI"A - 0
.
0
s so tsHEMOGOBIN 100 ml.
. 0
0
10 15
FIG. 2. PER CENT OF In CLEAREDFROMTHE MARROWPER MINUTE PLOTTEDAGAINST THE HEMOGLOBINCONCENTRATION
The relationship between marrow clearanceand muscle clearance is presented as a ratio ofmarrow K to muscle K (Table I). These ratiosare plotted against hemoglobin in Figure 3. Theratios of the non-leukemic, lymphoblastoma, andchronic myelocytic groups fall between 0.6 and1.4 (Figure 3). In the group of chronic lympho-cytic leukemia patients there was a scatter of theratio values from 0.80 to 4.0, with five patients be-low and four above 1.4. In the two patients withacute leukemia the ratios were 4.1 and 3.42. Thetwo patients with monocytic leukemia had ratiosof 1.03 and 0.28, and in three cases of multiple
myeloma, ratios of 1.67, 1.35, and 0.50 were ob-tained, respectively.
Although a wide distribution of Hb was pres-
ent in all groups of patients, there appears to beno direct relationship between the marrow tomuscle ratio and the hemoglobin concentration(Figure 3).
In a few patients it was possible to make ad-ditional clearance studies in order to ascertain thedegree of reproducibility of the technique in themarrow. The data obtained in four patients are
shown in Table III.
958
£CC13
a
I
-
"aU
5 10 15 5-tiEMOGLOIN GMS/I10O mi.
la 20-
I
5.
aw10-
a o
5-I'u
'uU.
BLOODFLOW IN HUMANBONE MARROW
DISCUSSION
The validity of the radioisotope clearancemethod as a measure of effective local circulatoryefficacy at the injected site was demonstratedtheoretically and experimentally by Kety (3, 4).Stone and Miller (13) experimentally demon-strated that the radioactive sodium (Na24) dis-appearing from the site of injection in the dogmuscle can be quantitatively recovered in thevenous blood draining the area of injection.Little or no radioactivity was recovered in thelymphatics draining the area. Subsequent studiesby others (4-10) have supported the contention
4.0-
'La,Qz4
.aIj
0
'hi
4.
'U
-I0LU
1.0
0
et
that locally injected radioisotopes can be used todetermine the effective local blood flow in muscleand skin utilizing Na24 and 18l. Minor unex-plained discrepancies have been detected betweenthe clearance rates obtained with these two iso-topes (8).
The extension of the Kety technique to arigidly enclosed and trabeculated organ such asthe bone marrow raises a question as to itsvalidity as an index of regional blood flow in thisorgan. Certain experimental conditions whichwould be expected to affect the systemic bloodflow were employed to demonstrate the validity
ACUTE LEUKEMIA- MMULTIPLE MYELOMA-O
0
0O
0
10 15
CHIROIC LYMPHOCYfCLIEUKEMIA
I 0
.0
0 0*
CHRONIC WELOCYTICLEUKEM- MoNOCYTIC LEUKEMIA- 0
Qb * 0. .
0
I00 15, , . . -.
HEMOGLOBIN GMS/100m1.
FIG. 3. RATIO OF MARROWCLEARANCEK To MUSCLECLEARANCEK PLOTTED
AGAINST HEMOGLOBINCONCENTRATION
NONLEUKEMIC MALIGNANCII
eS0 0
& , , , lb, ,
5 0 C5 1HEMOGLOBIN GM$ 100 ml.
W
z
4IU
4.0
UlQ~I 1.0-Q
2.0-
959
I
I
tIi
I
N. L. PETRAKIS, S. P. MASOUREDIS, AND PATRICIA MILLER
TABLE III
Demonstrating reproducibility of bone marrowclearance constonts
Patient Date Site Marrow K
W. F. 9/6/51 Sternum 11.29/6/51 Sternum 12.2
R. G. 9/13/51 Sternum 11.19/14/51 Sternum 13.2
E. C. 7/23/51 Sternum 8.27/23/51 Tibia 8.3
T. H. 7/23/51 Tibia 18.29/ 6/51 Ilium 15.0
of this technique in bone marrow. Following theintravenous injection of hexamethonium bromide,a ganglionic blocking agent, both the marrow andthe muscle clearance rates became very slow andappeared to be directly related to the pronouncedfall in blood pressure which resulted (Figure 4).Conversely, the injection of Paredrine resulted ina distinct increase in the clearance rate from themarrow, presumably as a result of the sympa-thomimetic action of this drug. These prelimi-nary experiments indicated that alterations in thesystemic circulation are reflected in the I"' clear-ance rate from the bone marrow. It is thereforefelt that the alterations in the bone marrow clear-ance rate produced by these pharmacologic agentsfurther support the interpretation that the clear-ance technique is a measure of effective bloodflow.
Of the many factors which could conceivablyinfluence the clearance rate of I18l in the marrow,the volume of fluid injected is an important con-sideration. The sudden injection of a volume offluid could possibly produce a local increase intissue pressure which in turn might increase theinitial rate of removal. With respect to bonemarrow, the 18l was injected slowly, withoutpain, and a period of 1 to 2 minutes elapsed be-fore counting was begun. This interval of timeallowed decompression of the injected site tooccur. In many of the cases, especially in thosepatients with acute leukemia and chronic lympho-cytic leukemia, the injected 1181 was literally as-pirated into the marrow, indicating that no pres-sure developed from the injection. The role ofthe lymphatic drainage of Il81 in the marrow isdifficult to evaluate. However it is probably negli-
gible, since no anatomic evidence for a lymphaticsystem in this organ has been presented (14).
Table III shows the degree of reproducibilityof the marrow clearance values in the same pa-tient on different occasions. Despite the technicaldifficulties encountered in making marrow clear-ance studies, it can be seen that a good agreementis obtained on repeated determinations.
The vascularity of the bone marrow is well rec-ognized. It was demonstrated by Tocantins,O'Neill, and others that many substances, in-cluding blood, plasma, medications, and diagnosticagents, when injected into the marrow cavityrapidly appear in the systemic circulation (15, 16).Tocantins reported that fluids could be adminis-tered at a rate of from 0.4 cc. to 25 cc. per minuteinto the marrow, but this rate probably does not re-flect the peak capacity for the removal of solutionsinjected into the marrow cavity because of changesin intramedullary pressure and physical disruptionof the marrow vascular bed. These factors as wellas the resistance to flow provided by the tubing,needle, and clotting of marrow blood probablyexert a marked and variable effect on the rate offlow of substances introduced into the marrow.Jones (2) has egtimated the rate of human hemato-poietic marrow flow to be approxirmately 300 cc.per min. based on data obtained on rabbits inradioactive colloid uptake studies by the marrow.
The data obtained in the present study indicatethat the clearance rates of 18l in such divergenttypes of vascular beds as bone marrow and restingmuscle are essentially identical in patients withvarious malignancies, Hodgkin's disease andlymphosarcoma. These rates are considerablyless than those reported by Wechsler, Sokoloff,and Kety (6) for human liver, where 34 to 48 percent of the injected isotope was removed perminute. Striking increases were observed in the1181 clearance rates in the bone marrow in leukemicpatients. The rapid clearance of radioiodide fromthe marrow in all the cases of acute leukemia, inhalf of those with chronic lymphatic leukemia, andin two of the three patients with multiple myelomasuggests that the marrow vascularity in these con-ditions is abnormal. In the patients with chronicmyelocytic and monocytic leukemia studied, theclearance rates were similar to those found in thenon-leukemic group of patients.
In an attempt to dissociate the systemic circula-
960
BLOODFLOW IN HUMANBONE MARROW
120
100
90
80
70
60a
0500.
co
z0
>
40x
40
30
20
5 10 15 5 10 15TIE IN MINUTES
FIG. 4. ALTERATIONS PRODUCEDIN I' CLEARANCEFROM THE MARROWFOL-LOWINGTHE INJECTION OF BISTRIUM AND PAREDRINE
tory influence on the local bone marrow flow, thedata were expressed as a ratio of marrow K tomuscle K. This ratio should serve to minimize theeffect of any generalized increase in the cardiacoutput on the marrow clearance rate, such as mightoccur in exercise, emotional stress, anemia, etc.It will be noted that in all the patients studied,with the exception of those patients with acuteleukemia, chronic lymphatic leukemia, and mye-loma, the ratio of marrow K to muscle K falls be-low 1.4. It is felt that a clearly demonstrablespecific local increase in marrow blood flow ispresent in those diseases with ratios above 1.4.
The demonstration of increased I'll clearance
rates in the bone marrow in leukemia might be an-ticipated since these diseases have been consideredto be neoplasms of the hematopoietic system (17).Many types of rapidly proliferating tumors havebeen demonstrated to elicit the formation of a vas-cular stroma (18, 19). The marked divergencebetween clearance rate found in those patients withacute and lymphocytic leukemia, and those withmyelocytic and monocytic leukemia is of interest.The present findings indicate that fundamentaldifferences may exist in the anatomy and physi-ology of the marrow vascular systems among thevarious human leukemias.
There is a lack of anatomical data on the vas-
961
I R
N. L. PETRAKIS, S. P. MASOUREDIS, AND PATRICIA MILLER
cularity of both normal and pathological humanbone marrow. A number of investigators (15, 20,21) have attempted to visualize the sternal mar-row vascular bed by injection of radiopaque sub-stances, but these studies were unsatisfactory forthe finer details of vascularity. No studies similarto those of Doan (22) in the pigeon were foundfor man. In observations made on histologicalsections of post-mortem bone marrow by Iwao (23)the marrow lesions in lymphocytic leukemia ap-peared to be associated with an increase in num-ber of capillary sinuses, whereas no apparentchange from the normal degree of vascularity wasobserved in the marrow in chronic myelocyticleukemia. Schleicher has also observed an in-crease in the degree of vascularity in the marrowassociated with the patchy and generalized marrowlesions in chronic lymphocytic leukemia.5 Suchanatomic differences in the vascularity of the mar-row may in part explain the findings obtained in thepresent study. Further detailed anatomic studiesare needed with regard to these observations.
The bone marrow cytology and peripheral bloodcounts could not be related to the increased clear-ance rates, other than in establishing the diagnoses.While blast cells were present in the marrow in allcases of acute leukemia and appeared to be posi-tively correlated with the rapid clearance ratesfound in these conditions, few or no lymphoblastswere observed in the chronic lymphocytic leukemiamarrows, yet very rapid clearance rates were ob-tained. In one patient with multiple myeloma(V. W.) almost no cells of any type could be as-pirated, yet the clearance rate was very rapid; 16per cent per minute. These findings suggest thatthe nature of the vascular bed may be more im-portant in determining the clearance rate ratherthan the presence or absence of abnormal cells inthe marrow.
Although most of the patients studied hadchronic anemia, no direct relationship was foundbetween the clearance rate and hemoglobin level(Figures 2 and 3).
These preliminary observations give promisethat the clearance rates of radioisotopes from themarrow may provide a means for the evaluation ofsystemic and local influences on the marrowcirculation.
6 Personal communication to authors.
SUMMARYANDCONCLUSIONS
The Kety radioisotope clearance technique withIl81 has been applied to a study of bone marrowand muscle blood flow in 43 patients with variousneoplastic diseases, lymphoblastomas, and leuke-mias. The marrow clearance rates were found tobe approximately the same as resting muscle in all17 of the non-leukemic patients studied.
In five cases of acute leukemia, marrow clearancerates were increased, being two to three timesgreater than the rate observed in the non-leukemicgroup, whereas, in five out of seven patients withchronic myelocytic leukemia, and in two cases ofmonocytic leukemia, the marrow clearance rateswere within the range of the non-leukemic groupof patients. In six out of ten patients with chroniclymphocytic leukemia, elevated clearance rateswere obtained. Three out of four patients withmultiple myeloma had increased marrow clearancerates. When marrow K to muscle K ratios wereobtained in the non-leukemic, chronic myelocyticand monocytic leukemic patients, the values fellbelow 1.4. In two cases of acute leukemia andfour out of ten chronic lymphocytic leukemics theratio obtained was well above .1.4. The findingsappear to demonstrate a selective increase in bloodflow in the bone marrow in acute leukemia and insome cases of chronic lymphocytic leukemia andmultiple myeloma, and strongly suggest that funda-mental differences in the marrow vascular bedexist among the leukemic group of diseases.
Alterations were produced in the clearance ratesin bone marrow following the intravenous injec-tion of Paredrine and hexamethonium in two pa-tients. The radioisotope clearance techniquepromises to be a useful tool for the investigationof the pathologic-physiology of the human bonemarrow circulation.
ACKNOWLEDGMENT
We wish to express our appreciation to Dr. JonahG. Li of the Division of Preventive Medicine, Universityof California School of Medicine, San Francisco for hiscooperation in providing many of the patients for thesestudies.
REFERENCES
1. Green, H. D., Methods in Medical Research. Vol. I,Section II, V. R. Potter, Ed., Year Book Pub-lishers, Inc., Chicago, 1948.
962
BLOOD FLOW IN HUMANBONE MARROW
2. Jones, H. B., Medical Physics. Vol. II, OttoGlasser, Ed., Year Book Publishers, Chicago,1950, p. 860.
3. Kety, S., S., Quantitative measurement of regionalcirculation by the clearance of radioactive sodium.(Proc. Physiol. Soc. Phila., Jan. 20, 1948), Am.J. M. Sc., 1948, 215, 352.
4. Kety, S. S., Measurement of regional circulation bythe local clearance of radioactive sodium. Am.Heart J., 1949, 38, 321.
5. Elkin, D. C., Cooper, F. W., Rohrer, R. H., Miller,W. B., Shea, P. C., and Dennis, E. W., The studyof peripheral vascular disease with radioactiveisotopes. Part I. Surg., Gynec. & Obst., 1948,87, 1.
6. Wechsler, R. L., Sokoloff, L., and Kety, S. S.,Measurement of hepatic circulation by clearanceof radioactive sodium in man. Federation Proc.,1950, 9, 133.
7. Franke, F. R., Boatman, J. B., George, R. S., andMoses, C., Effect of physical factors on radio-sodium clearance from subcutaneous and intra-muscular sites in animals. Proc. Soc. Exper.Biol. & Med., 1950, 74, 417.
8. Boatman, J. B., Kendrick, T. R., Franke, F. R., andMoses, C., The use of radioactive iodine, radio-active phosphorus, and radioactive sodium in thedetermination of cerebral and muscle clearance.J. Lab. & Clin. Med., 1950, 36, 456.
9. Wisham, L. R., Yalow, R. S., and Freund, A. J.,Consistency of clearance of radioactive sodiumfrom human muscle. Am. Heart J., 1951, 41, 810.
10. Rapaport, S. I., Saul, A., Hyman, C., and Morton,M. E., Tissue clearance as a measure of nutritiveblood flow and the effect of lumbar sympatheticblock on such measures in calf muscle. Circula-tion, 1952, 5, 594.
11. Leblond, C. P., Advances in Biological and MedicalPhysics. I. J. H. Lawrence & J. G. Hamilton,Ed., Academic Press, Inc., New York, 1948,p. 353.
12. Siri, W. E., Isotopic Tracers and Nuclear Radia-tions, McGraw-Hill Co., New York, 1949.
13. Stone, P. W., and Miller, W. B., Mobilization ofradioactive sodium from the gastrocnemius muscleof the dog. Proc. Soc. Exper. Biol. & Med., 1949,71, 529.
14. Drinker, C. K., and Yoffey, J. M., Lymphatics,Lymph and Lymphoid Tissue, Harvard UniversityPress, Cambridge, Mass., 1941.
15. Tocantins, L. M., Rapid absorption of substancesinjected into the bone marrow. Proc. Soc. Exper.Biol. & Med., 1940, 45, 292.
16. Tocantins, L. M., and O'Neill, J. F., Infusions ofblood and other fluids into the general circulationvia the bone marrow. Surg., Gynec. & Obst.,1941, 73, 281.
17. Willis, R. A., Pathology of Tumors, C. V. MosbyCo., St. Louis, 1948.
18. Algire, G. H., and Chalkley, H. W., Vascular re-actions of normal and malignant tissues in vivo.I. Vascular reaction of mice to wounds and tonormal and neoplastic transplants. J. Nat. CancerInst., 1945, 6, 73.
19. Bierman, H. R., Byron, R. L., Jr., Kelly, K. H., andGrady, A., Studies on the blood supply of tumorsin man. III. Vascular patterns in the liver byhepatic arteriography in vivo. J. Nat. CancerInst., 1951, 12, 107.
20. Benda, R., Orinstein, E., and Depitre, M., Injectionsintramedulaires osseuses de substances opaqueschez l'homme. (Atude comparee des imagesradiologiques obtenues sur les pieces anatomiqueset sur le vivant.) Sang., 1940, 14, 172.
21. Pizzolato, P., The blood supply of the sternum. I.X-ray studies of injected sternums showingvenous return. New Orleans M. & S. J., 1944-1945, 97, 71.
22. Doan, C. A., The circulation of the bone marrow.Contrib. Embryol., 1922, 14, 27.
23. Iwao, T., 'Ober die Knochenmarksbefunde bei Fallenvon verschiedenen Blutkrankheiten, besonders uberden Zusammenhang zwischen der Blutbildung unddem Entwicklungsgrad der Kapillarsinus imKnochenmark. Transactiones Societates Patho-logical Japonica, 1940, 30, 54.
963
