Musculoskeletal Susan

K. Stevens,

MD

#{149} Sheila

Repopulation MR Imaging

G. Moore,

terms:

331.341,

Bone

331.342

331.342

marrow,

B

#{149} Leukemia,

ic resonance Spine.

(MR),

MR

studies,

Radiology

disease,

therapy, tissue

imaging

vasive

Transplantation: Correlation’

is an effective,

modality

that

logic changes we instituted determine regeneration spine after

appearance on MR images transplantation.

PATIENTS

AND

years) who (six patients)

marrow

in

were MR

were

ages 300

Magnet-

#{149}

were msec)

of the

15 patients

transplants resistive

images

METHODS

received either or allogeneic

sonex, Sunnyvale, echo (SE) 300/30, msec/echo time 75

of marrow of the

MR examinations

spine

0.38-T

patho-

the

Sixty-seven lumbar

nonin-

reveals

in bone marrow (4-14), a prospective study to

system

(aged

3-44

autologous (nine patients)

performed (RX

on a

4000;

Re-

Calif). Sagittal spin75 (repetition time [TR] [TE] msec) and 2,000/30, obtained with a body coil

and iO mm-thick between consecutive

characterization

MD

ONE

(MR)

therapy.

331.341

331.1214,

D. Amylon,

marrow transplantation has been performed successfully since the mid-i950s. The diseases most amenable to transplantation are hematobogic malignancies; however, nonneoplastic disorders, including immune deficiency diseases, inborn errors of metabolism, and severe madiation injury, are also treated succcssfulby with transplantation (1-3). Currently, bone marrow transplants are assessed with serial bone marrow biopsies and aspirations performed in conjunction with peripheral blood tests. Because magnetic resonance

transplantation,

#{149} Hodgkin

#{149} Michael

of Marrow after with Pathologic

Sixty-seven magnetic resonance (MR) studies of the lumbar spine were performed in 15 patients with bone marrow transplants, and the appearance of marrow regeneration on MR images was correlated with results of bone marrow biopsy and pathologic examination. After transplantation, Ti-weighted MR images of vertebral marrow showed a charactenistic band pattern consisting of a peripheral zone of intermediate signal intensity and a central zone of bright signal intensity. Reciprocal changes were identified on short inversion time inversion recovery images. At histologic examination the central zone corresponded to fatty marrow; the peripheral zone corresponded to a zone of regenerating hematopoietic cells. Posttransplantation Ti and T2 relaxation times of the entire vertebral marrow were calculated from the spin-echo images; no statistically significant trends in relaxation times were noted. Knowledge of the normal MR pattern of marrow regeneration after transplantation may be useful in screening for residual marrow disease, determining marrow engraftment, and differentiating marrow repopulation with normal versus malignant cells. Index

MD

sections with sections.

obtained images

a 20% gap Ten aver-

for TI-weighted and

two

(TR

averages

were

obtained for 12-weighted (TR 2,000 msec) images with a 256 X 256 image tnix. Short inversion time (TI) inversion recovery (STIR) images were acquired

.

331.92

1990; 175:213-218

with

a TR

of

1,500

msec,

ma-

TE of 30 msec,

and TI of iOO msec with two averages per phase-encoding step. (A TI of 100 msec

I

Felix

From

the

Bloch

058A,

Departments

Laboratory

Stanford,

sion

requested

part

by

the

RSNA,

CA

of Diagnostic

for 94305-5105.

September Mellon 1990

Research

Foundation,

Radiology

(S.K.S.,

From

6; revision Stanford

(S.K.S.,

S.C.M.),

the

1988

received

RSNA

20; (S.C.M.).

and

University

annual

November

University

S.C.M.)

Stanford

Pediatrics

Medical

meeting.

Received

accepted

December

Address

reprint

Radiology

(M.D.A.)

and

Center, July

to suppress

to fat

and

marrow

lowing tissues with times to demonstrate tensity.)

the signal

fatty

Patients

intensi-

while

al-

longer Ti relaxation increased signal in-

were

admitted

to the

hospital and an initial MR study of the lumbar spine was performed at least 8 days before transplantation. All patients then received a pretransplant preparatory regimen consisting of either chemotherapy

alone

tion tal

(seven

patients)

of chemotherapy body

radiation

or a combina-

and (eight

fractionated patients)

to(Table).

Patients were imaged immediately after the preparatory regimen but before transplantation, and then at regular intervals (day 0-40, day 40-90, day 90-i80, and after day i80) up to i4 months after transplantation. Ti relaxation times were calculated with successive approximation to fit the acquired data to an exponential curve. 12 relaxation time was calculated with the signal intensity measurements for two different TEs and the assumption that hydrogen density, TR, and Ti are constants. Signal strength is therefore proportional to exp(-TE/T2). The regions of interest used in calculating the relaxation times included vertebra! marrow from L-l through L-5 demonstrated on any particular image, provided the vertebral body was not near the edge of the coil. Regions of interest were drawn to include as much marrow for each vertebral body as possible, with care taken not to include cortical bone, disk, the basivertebral venous plexus, or postenor spinous ligaments. In each case, the individual regions of interest consisted of more than 40 pixels, with an average standard deviation of calculated relaxation times of less than iO%. There are known problems with accuracy of a twopoint data fit method for calculation of relaxation time (4), and in light of this, the relaxation time results must be viewed with caution. A linear model incorporating the method of least squares was used to analyze the calculated Ti and T2 relaxation times as a function of time

5-

revi-

6. Supported requests

ty due

the

Room

5, 1989;

was chosen

to S.C.M.

in

Abbreviations: TI

inversion

version

SE recovery,

time,

TR

spin TE repetition

echo,

STIR

echo

time,

short TI

in-

time.

213

following nificance propriate

transplantation. was determined F test.

MR images

were

Statistical using the

interpreted

sig ap-

by two in-

dependent observers experienced in the interpretation of MR images of marrow. Images were interpreted without knowledge of transplant status (whether pre- or posttransplantation) and without knowledge of the time elapsed since transplantation

in

each

patient.

Observers

were

also blinded to clinical status and marrow aspiration or biopsy results obtained at the time of the MR examination. Marrow signal intensity less than that of muscle

was considered low signal mow signal intensity equal

intensity, marto or slightly

greater than that of muscle was considered intermediate signal intensity, and marrow signal intensity slightly less than or equal to that of fat was considered increased signal intensity. A pattern of central increased signal intensity and periphemal intermediate signal intensity was seen during data analysis; because this pattern often had the appearance of alternating bands of differing marrow signal intensity, this was termed the “band pattern.” Bone marrow biopsies and aspirations of the posterior iliac crest were usually performed within 48 hours of the MR

examination

during

the first 3 months

after transplantation and within 2-4 weeks of the MR examination during months 4-i4. The results of penipheral blood smears and bone marrow biopsies were reviewed and correlated with the MR results only after the

MR images

were

acquired

and

a.

inter-

TE

preted.

RESULTS

nized

in the

Table.

The

mar-

of the

tients in disease remission demonstrated intermediate

pre-

is summa-

marrow

zone

of pa-

generally signal

signal

intensity

and T2-wcightcd tients (one with transplant)

on

a peripheral

zone

signal intensity of increased signal

and in-

tensity. Postpreparatory

pretransplant marpatients were imaged immediately following the pmepamatory regimen but prior to tmanspbantation. The marrow pattern in these patients showed a slight decrease in signal intensity on the Ti-weight-

row-Four

ed images

as compared

mow pattern image. 214

#{149} Radiology

of the

with

prcpmepamatory

the

mar-

shows

reciprocal

of bright

signal

marrow

intensity

90 days after transplantation, some individuals manifested tern as early as 40 days after plantation (Fig 2).

Ti-

images. Three paa prior bone marrow

showed

of intermediate central zone

both

msec])

on the

changes.

0 U,

Ti-weighted images (Fig 1). We have termed these alternating zones of intemmediate and bright signal intensity the band pattern. The band pattern was detected in all but one patient by

in-

tensity on Ti- and 12-weighted images. The marrow of patients in disease relapse demonstrated abnormally low

msec/TI

Posttransplant marrow-Characteristic marrow changes were seen on the SE images within 3 months after transplantation. These consisted of a peripheral zone of intermediate signab intensity surrounding a central

SE Imaging Prepreparatory pretransplant row.-The MR appearance transplant marrow images

b.

Figure 1. Sagittal MR images of lumbar spine obtained on day 156 after transplantation. (a) Ti-weighted SE 300/30 image. Note peripheral zone of intermediate signal intensity surrounding a central zone of bright signal intensity. (b) STIR image (1,500/30/ 100 [TR msec/

a

A single patient initially manifested

and this trans-

pat-

intensity

Bone

on

marrow

the Ti-weighted biopsy the

day revealed acute leukemia in relapse

Figure

2.

ing

early

an

pattern

whose marrow the band pattern

demonstrated abnormally

diffusely low signal images. following

nonlymphocytic (Fig 3).

40-90 DAYS

in the first few months after transplantation developed a diffusely homogeneous marrow pattern by 8 months after tmansplantion. Followup MR examination performed 6

months later homogeneous,

16-40

Number

the

tion

of

band

after

far

lapse an

of initial

in disease

STIR

(hatched

the during

band

homogeneous

marrow

those

demonstratbar)

as

transplantation.

right (late graphically

examinations

demonstratbone

versus

pattern

time

bar on the row pattern)

of studies

bar)

180+

#{149}

TRANSPLANTATION

homogeneous

(blank

ing

#{149} 90-180

POST

homogeneous represents

only

a func-

The

pattern

evolved

marrow

pattern

marMR

two

patient the

solid

with

study

in into with

rewhich a more time.

Imaging

Prepreparatory

pretranspiant

row-Marrow

of

remission

generally

patients

showed

marin

disease

deApril

1990

Pretranspiantation

Clinical

and MR Imaging

Data MR Imaging

Age (y)

Patient 1

Pretransplantation

10

2

18

3

19

4

Marrow

Diagnosis Non-Hodgkin lymphoma

mia Non-Hodgkin

lymphoma Acute nonlymphocytic mia

5

13

Acute

Remission

sulfan Etoposidet

Remission

FTBI Cyclophosphamide sulfan

Relapse

and

Etoposidet

leukemia

Cyclophosphamide and busulfan

7

i5

Non-Hodgkin lymphoma

Remission

8

12

Acute nonlymphocytic leuke-

Remission

mia Hodgkin

disease

41

Acute

Remission

nonlymleuke-

SI

creased

Peripheral: creased

NA

NA

Homogeneous,

Technically

sub-

intermediate SI Peripheral: inter-

optimal Peripheral:

in-

SI; bright

mediate central: creased

Homogeneous,

creased SI; central: low SI

SI; inSI

Homogeneous,

SI

Homogeneous,

intermediate

SI

Cyclophosphamide, etopo-

Homogeneous, low SI

side,t and FTBI Etoposidet and FTBI

Homogeneous, intermediate

Homogeneous, intermediate

SI

Peripheral: creased

BCNU,

Homogeneous,

SI

low

etopoand

cy-

SI

Peripheral: increased SI; central: low SI

SI

intermediate

Homogeneous, intermediate

SI

Homogeneous, intermediate

SI

Cyclophosphamide, etopo-

Homogeneous, intermediate

SI

Homogeneous, intermediate

SI

Homogeneous, intermediate

SI

Homogeneous, intermediate

SI

in-

SI; central: low SI Homogeneous,

Homogeneous,

intermediate

Remission

brightS!

Peripheral: intermediate SI; central: increased SI Homogeneous, intermediate SI

clophosphamide Etoposidet and busulfan

phocytic

in-

SI; central: low SI

inSI

Peripheral: intermediate SI; central: bright SI Homogeneous, intermediate SI

side,t

10

creased

low

Remission

24

central:

mediate central: SI

and

interSI;

in-

intermediate SI Peripheral: inter-

FIB!

Acute lymphocytic leukemia

Peripheral: mediate

STIR Image

Homogeneous,

bu-

15

interSI;

central: NA

and

6

9

Peripheral: mediate

FTBI

Cyclophosphamide and bu-

leuke-

lympho-

cytic

Image

Cyclophosphamide and

Remission

T2-weighted Image

Tl-weighted

Regimen

Remission

Acute nonlymphocytic leuke-

7

Preparatory

Status

Appearance

SI

Homogeneous, low SI

mia

ii

23

Hodgkin

disease

Remission

side,t

i2

28

13*

Hodgkin

4

14

-Thalassemia

19

15

Remission

disease

28

Relapse

Acute

Remission

nonlym-

phocytic

Etoposidet FTBI

leuke-

Homogeneous, low SI Homogeneous, low SI

and

NA

Homogeneous, intermediate Homogeneous, intermediate

Homogeneous,

NA SI NA SI

Homogeneous,

intermediate

SI

in-

creased SI; central: low SI

FTBI

Cyclophosphamide, etoposide,t anti BCNU Cyclophosphamide and FTBI Cyclophosphamide and busulfan

Relapse

Acute nonlymphocytic leukemia

and

Peripheral:

NA

intermediate

SI

mia

Note.-Cyclophosmamide nitrosourea)

signal

New

York);

intensity.

The

intensity

on

the

STIR images. Results arc summarized in the Table. Postpreparatory pretransplant marrow.-Onby

one

STIR imaging preparatory

immediately regimen but

patient

compared

with

preparatory Posttransplant

whose

marrow

Volume

175

The

dose:

rad

[13.2

four

Gyj),

times

daily

for 4 days;

intensity,

NA

BCNU not

(1,3-Bis(2-chlor08thyl)-1-

available.

band

marrow showed signal intensity

pre-

demonstrated 1

patients the

on the

intensity alterzone of bow sigthe was the

SE images

a trend

toward

in the

calculated

Ti

of the pbetion

vertebral of the

marrow preparatory

of the

increase

relaxation

but prior to transplantation. may reflect early cellular

edema

No

discernible

time

was

Pathologic

Times was

calculated Ti relaxation time of the vertebral marrow during the 2 months after marrow tmansplantation, followed by a subsequent Ti prolongation; however, these findings were not statistically significant.

marrow

radiation and chemotherapy. was a trend toward decrease

trend

seen

(Fig

in T2 relaxation

4).

was

discerned.

Relaxation

and

image. marrow-All

pattern

initially

There

after the prior to

on the

60 mg/kg.

zone of bright signal nating with a central

underwent

marrow

#{149} Number

(1,320

1 mg/kg

nab intensity (Fig i). In all cases band pattern on the STIR images identified at the same time that

signal

transplantation. diffusely increased

irradiation

transplant. Bristol-Myers,

marrow of the single patient who was imaged with STIR sequences while in disease relapse showed difbright

body

dose:

tral

fusely

total

busulfan

SE on

of low

fractionated

for 2 days;

band pattern on the Ti-weighted images showed reciprocal changes the STIR images, that is, a peripheral

(Vepesid;

FTBI

daily

creased signal intensity on the STIR images. Five patients (one with a pnior bone marrow transplant) demonstrated a peripheral zone of increased signal intensity surrounding a cen-

bone marrow

t VP-16-213

zone

mg/kg.

once

signal

Second

5-15

dose: 60 mg/kg

SI

C

dose:

time

after cornregimen This necrosis

caused Theme in the

by

Correlation

Aspirates of iliac crest bone marrow from all patients demonstrated successful engraftrnent of erythroid and rnyeloid elements. Pathologic cxamination of the iliac crest marrow revealed hypocellulanity ranging from 5% to 30%, with varying proportions of fat, fibrosis, or both. A vertebral body specimen was obtamed from a patient who died of a ruptured

Meckel

diverticulum

Radiology

on #{149} 215

b.

a. Figure row in signal neous,

day

3i after

transplantation,

and

the

gross specimen was examined histologically (Fig 5). Immediately adjacent to the cartilaginous end plate, a hypercebbulam zone of mepopubating

hcmatopoietic that

cells

corresponded

was

identified

to the

peripheral

zone of bow signal on the Ti-weighted

intensity image.

observed The ccn-

tral

marrow

contained

portion

of the

bone trabeculae with a barge of marrow fat and a relative of hematopoietic precursors.

amount paucity

DISCUSSION Bone coming peutic

marrow transplantation is beincreasingly used as a themamodality

for

the

early

treat-

ment of malignant hematobogic orders. Early recognition of cngraftmcnt with clinical blood smears,

Marrow

216

marrow

undergoes

stresses. .

dis-

is currently assessed criteria, peripheral and marrow biopsy.

in bone

recipients

logic

c.

3. Sagittal SE 300/30 images. (a) Vertebral marrow on day 180 after transplantation shows typical band pattern. (b) Vertebral marsame patient on day 250 after transplantation shows development of more homogeneous pattern as peripheral zone of intermediate slowly expands. (c) Vertebral marrow in the same patient i4 months after transplantation. Marrow demonstrates diffusely homogeabnormally low signal intensity. Biopsy showed acute nonlymphocytic leukemia in relapse.

Radiology

The

transplant unique

large

doses

physio-

of che-

mothemapy and fractionated totalbody radiation administered in the pmetransplant preparatory regimen arc designed not only to induce immunobogic suppression in the recipient but also to eliminate any residual malignant cell populations. After pmctranspbant chemotherapy and ma-

signal

diation, bone marrow cells are infused intravenously and the stem cells “home” to the marrow cavity aften a transient residence in the lungs and spleen. Hematobogic engmaftment typically takes 3-4 weeks and is heralded by a peripheral rise in gmanubocytes (2,3,15). Our results show a characteristic

the

vertebral

the

unique

MR

change

after

bone

in the marrow

this change consists zone of intermediate

vertebral

marrow

transplantation;

of a peripheral signal intensity

intensity

STIR tion,

are

images. At the peripheral

sponds

to a concentrated

populating

cells,

of the

body

(Fig

is

of

a result

to the

of verte-

6). bone

marrow,

is concentrated beneath the bone

he-

in the cortex,

whereas the central portion of the marrow cavity remains relatively fatty. The vascular sinusoids control blood flow through the marrow cavity and determine the size of the hematopoietic compartment (17). Blood flow to the vascular sinusoids, in

osteal capillaries, into the marrow

signal intensity suncentmal zone of decreased

as

flow

In hematopoietic

matopoiesis periphery

of mecentral of pattern

mepopubation

marrow

blood

zone

a

the

a reflection

turn,

of bright

zone

whereas

a preponderance We believe this

bral

on the examinacome-

zone reflects marrow fat.

surrounding a central zone of melatively bright signal intensity on the Ti-weighted images. Reciprocal changes consisting of a peripheral rounding

identified

histologic region

nutrient

is carried arteries

by branches

of the

that

terminate in capillaries near the endosteab surface of the cortex, anastomose with pen-

and then turn back to empty into sinusApril

1990

U 0) Cl)

E I-

I

I

I

0

I

I

#{149}

Days

I

after

I

I

I

I

.

.:

0

Transplantation

I

I

I

I

.

Days

after

Transplantation

b. Figure

4.

Mean

Ti (a) and T2 (b) relaxation

times

before

and after

transplantation.

b. Figure 5. adjacent to of marrow generating

C.

(a) Photomicrograph of pathologic specimen from vertebral body shows a hypercellular zone the cartilaginous end plate (bottom). The central portion of the vertebral body (top) contains fat (original magnification, XiOO). (b, c) Higher magnification views (original magnification, cells adjacent to end plate (bottom) and (c) central marrow fat surrounding trabecular bone.

generation seen on MR images in our transplant patients. Although the temporal evolution of this pattern is variable, it was seen in all but one of our marrow transplant patients. The appearance of the vertebral marrow on MR images obtained after the preparatory regimen and in the initial

Figure 6. Arterial injection demonstrates the anatomy of the arterial blood supply to vertebral marrow. Terminal branches of the nutrient artery end in capillaries near the endosteal surface of the cortex just beneath the

end

from

plate.

reference

(Reprinted,

with

permission,

i6.)

transplantation

tion changes in marrow consist of death of radiation-sensitive cells within the first few hours of exposure, with subsequent congestion, disruption of marrow sinusoids, edema, and hemorrhage of the marrow

and

subsequently

and

the

basiventebrab

vein

The

ramifications

during the next 5 days. tant increase in fatbike

in the

vertebral

noted.

This

me-

zation

of the

of the body

nutrient occur

gion

just

or end

terminal

artery in the

beneath

plate.

We

vertebral the band

Volume

175

(Fig

metaphyseal

the

fused

believe

this

of blood flow determines bution of repopulating the for

venubes

after

oids (i7-i9).

into

weeks

may reflect a number of hemodynamic and physiologic phenomena attributable to the combined effects of substantial radiation-induced mannow necrosis and early hematopoictic reconstitution. In mats, initial madia-

body pattern #{149} Number

6)

epiphysis pattern

the distnicells within

and is responsible of marrow mc1

A concomimaterial is

is followed

marrow

by

strorna

and

re-

establishment of the integrity of the vascular sinusoids, then by repopulation with foci of primitive hemato-

poietic tion

cells (20).

underwent

that

In some imaging

undergo of our

regenemapatients

immediately

ten the

brab

who af-

hematopoietic

preparatory

to transplantation, crease in signal

cells

marrow

regimen

the intensity on

the

but

prior

initial deof the vente-

Ti-weighted

irn-

ages, the corresponding increase in signal intensity on the STIR images, and the initial increase in the calcubated Ti relaxation time may reflect early radiation-induced marrow edema and necrosis. The development of the band pattern in our patients in the weeks aften transplantation reflects early hematopoietic ally within

reconstitution the vertebral

penipherbody,

whereas the central portion of the vertebral body retains a barge amount of marrow fat. The band pattern persisted

in all

the

patients

we

studied

except one who had a disease after transplantation. In this uab, we observed the evolution initial

neorgani-

of repopulating

bone trabeculae and a large amount X400) of (b) peripheral zone of re-

posttmansplantation

tern

into

This

homogeneous

eventually

a more

relapse individof the

band

homogeneous marrow

developed

pat-

one. pattern

a diffuse,

ab-

normally low signal intensity with an abnormally prolonged Ti relaxation time, and the patient relapsed

i4 months them

after

longitudinal

transplantation. studies

must Radiology

Furbe

217

#{149}

done to determine whether the yentebmal marrow in normal posttnans-

the

plant

Although

patients

continues

to exhibit

the band pattern indefinitely, or whether this band pattern will change and become more homogeneous with time. Both the appearance on MR images and the relaxation

time

may

ultimately

be

toward

initial

could

reflect

transient

months

after

time

than

and

transplantation

yellow

marrow,

in Ti relaxation hematopoictic and have not been shown

tween

row

tistically

significant

(22).

on

the

imaging

transplant there was

edge

explained

by

light

marrow

10.

1 1.

using

transin

logic

analysis;

to Karl

Blume,

able contribution; and for his kind assistance

our

MD,

deep his

for

ap-

histo-

his

valu-

to Mark Riesenberger in preparing the manu-

16.

18.

ED,

marrow 2.

Storb

R, Clift

transplantation. 292:832-843,

Gale

RP.

296. Thomas

Bone

tion.

ED. In:

tersdorf AS,

eds.

4.

Vogler

assessment son Imaging

5.

6.

marrow

transplanta-

Braunwald

E, Isselbacher

RC,

JD,

Wilson

Harrison’s ilth

ed.

Protocols

Sale

Pe-

before,

21.

equipment. 6:195-199.

Magn

the

during,

and

after

treat-

ma. Invest

bone aplastic

Radiol

marrow anemia,

1986;

in

patients and

22.

imaging

Bone

of and

marrow

168:679-693. marrow

In:

Sale

with CE,

vertebral

Shul-

Orthop and

[Br]

Clin

of bone

1967; 52:13-23. The vascular anato-

J.

Trueta spine

Surg

Amato

column.

115:6-21. histophysiology

vertebral

its

relationship

to

J Bone

osteomyelitis. 1959;

VP,

Bombelli

cular supply the growing

of the rabbit.

41:796-809.

R.

The

vertebral

J Bone

normal

vas-

column

in

Joint

Surg

[Br]

41:782-795.

Rubin P. Casarett sues and blood. Vol

In:

CW. Hematopoietic Clinical radiation

2. Philadelphia:

tispa-

Saunders,

778-849.

Dooms

CC, M,

Fisher

Crooks

marrow

Re-

Radiology 1987; 162:701-707. DA, Shields AF, Sheurich CJ, Porter BA, Moss AA. Magnetic resonance imagof the

AM,

son for

WA.

lumbar

of the

1968;

de Certaines

MR

system: current 1987; 149:457-467.

of bone

Clin

Wiley

thology.

objects

D.

Pathology

1976; L. The

McGraw-

test

Resnick

1988;

of the

Orthop Weiss

of internal

DW,

1984;

man HM, eds. The pathology of bone marrow transplantation. New York: Masson, 1984; 215-230. Crock HV, Yoshizawa H. The blood sup-

principles

ment. Olson

leukemia,

20.

JB, Fauci

York:

R, et al. of bone mar-

Radiology

transplantation.

1959;

KJ,

Martin

146:353-358.

Radiology

CE.

New

and

of MRI 1988;

2:261-

McKinstry CS, Steiner RE, Young AT, Jones L, Swirsky D, Aber V. Bone marrow in leukemia and aplastic anemia: MR imaging

a prior

Bone

19.

1985;

RT, et al.

in children.

EC, Baehner imaging

JB, Murphy

pyogenic

transplantation.

Hematol

1986;

RB, Tiara

in children.

DJ,

Joint

marrow

Am

marrow

1986;

Klatte resonance

disease

Sartoris

my

J Med

N Engl

Hill, 1987; 1536-1541. Lerski RA, McRobbie et al.

Bone

895-902.

CRC Crit Rev Oncol 3.

RA, et al.

AJR

MD,

marrow.

References Thomas

North

of bone

marrow.

Cohen Magnetic

ply

script.

1.

Magmarrow

Assist Tomogr 1986; 10:205-209. Daffner RH, Lupetin AR, Dash N, Deeb ZL, Sefczek RJ, Schapiro RL. MRI in the detection of malignant infiltration of

marrow for

im-

of bone

Clin

the musculoskeletal future status. AJR

15.

MD,

resonance

151:715-718.

repopulaversus

We express

deby

1987; 5:225-230. AF, Olson DA.

H, Dietrich

imaging

imaging.

Baum,

The

involvement

imaging

Radiol

Kangarloo

row

14.

to Charles

S. Olson

ED.

magnetic

resonance

bone 12.

ment, and differentiating tion of marrow with malignant cells. U

predation

Ra-

J Comput

may Knowl-

normal

Acta

Churchley

marrow

J Clin Oncol BA, Shields

MR

of man-

after useful

BA,

FR, Thomas

of bone

Porter

diseases.

24:269-289.

trans-

pattern

Acknowledgments:

Porter

disorders.

of the

13.

medicine.

in

AF,

Appelbaum

netic

screening for residual marrow disease, determining marrow cngmaft-

of

were repro-

in one patient; however, no prior transplantation

#{149} Radiology

9.

marrow

B, et al. in diffuse

28:199-205.

lymphoma

of evaluating

before and could prove

Shields

aging.

imaging

means

bone

tection

suppres-

marrow MR

normal

mow change plantation

ing

218

bone that

R, Rehn S. Climelins resonance imaging

1987;

DA,

vertebral

marrow.

of the

JD,

be

8.

determined

vertebral

posttnansplantation

results.

can

with

suggests

1975;

Prior to transplantation a band pattern was noted on the SE images from three patients and on the STIR images from five patients. These findings

in the

be a noninvasive

system

because the regions of interest not sufficiently large to allow ducible

plants

diol

mc-

to the

Nyman Magnetic malignant

be

transplantation, a more general

supply

of patients

7.

17.

these reports, the lack of statistical significance of our Ti relaxation data may not be unexpected. Relaxation times of the central and peripheral marrow zones could not be calcubated separately

may

of hematopoictic

pattern

mc-

the

time befatty manto be staIn

pattern

marrow reflects

blood

band

radiation-

differences

band

body after complete marrow sion. The consistent development

(Fig 4). This increase could reflect continuing hematopoictic meconstitution and expansion of the population of regenerating cells within the marnow cavity. However, while it has been suggested (2i) that hematopoietic marrow has a longer Ti relaxation

had

chemotherapy.

in a pattern

by the

useful

induced marrow necrosis and edema. This initial increase and subsequent decrease was followed by an appamcnt increase in the Ti relaxation time in the

the

constitution

increase

they

extensive

phenomenon

then decrease in the calculated Ti melaxation time of the vertebral marrow after completion of the preparatory regimen was seen in several patients. This

although

prior

specific for it probably

in differentiating a fully meconstituted, healthy posttransplant marrow from one that has relapsed. A trend

remainder,

ceived

MR.

LE,

imaging:

studies

related

1985;

155:429-432.

Jenkins

JPR,

Hickey

DS,

Quantitative

of vertebral Magn Reson

Hricak

Cenant magnetic

to age Stehling Hillier magnetic

RichardBone

resonance

and M,

VF,

H,

HK. sex.

Radiology

Sivewright

C,

Isherwood resonance

I. imaging

bodies: a Ti and 12 study. Imaging 1989; 7:17-23.

with lympho-

21:540-546.

April

1990