Journal Leiomyoma

8/11/2019 Journal Leiomyoma

1/913

ORIGINAL ARTICLE

Ann Nucl Med (2008) 22:803810

DOI 10.1007/s12149-008-0184-6

S. Nishizawa (*) M. Inubushi A. Kido M. Miyagawa T. Inoue K. Shinohara M. KajiharaHamamatsu Medical Imaging Center, Hamamatsu MedicalPhotonics Foundation, 5000 Hirakuchi, Hamakita-ku,Hamamatsu, Shizuoka 434-0041, Japane-mail: [email protected]

Incidence and characteristics of uterine leiomyomas with FDG uptake

Sadahiko Nishizawa Masayuki Inubushi Aki KidoMasao Miyagawa Takeshi Inoue Katsura Shinohara

Makoto Kajihara

12 examined more than twice showed substantial changes

in the level of FDG uptake in leiomyomas each year with

FDG uptake disappearing or newly appearing. These

changes were observed frequently in relation with meno-

pause or menstrual phases.

Conclusions Leiomyomas with focal FDG uptake were

seen in both pre- and post-MP women with a higher

incidence in pre-MP women. Abundant cellularity and

hormonal dependency may explain a part of the mecha-

nisms of FDG uptake in leiomyomas. It is important to

know that the level of FDG uptake in leiomyomas can

change and newly appearing FDG uptake does not nec-

essarily mean malignant transformation.

Keywords FDG-PET MRI Uterine leiomyomas

Genitourinary oncology

Introduction

Positron emission tomography (PET) using 18F-fluoro-

deoxyglucose (FDG) has been proved to be an effective

diagnostic tool for a variety of malignant tumors and is

frequently used for the management of patients with

such tumors. However, it is true that many benigntumors and diseases or other physiological conditions

also show focal FDG uptake that mimics that of malig-

nant lesions and leads to misinterpretation of FDG-PET

images [13]. Therefore, it is important to understand

those conditions as much as possible to prevent misin-

terpretation. Recent articles showed that, as diagnostic

pitfalls specific to the pelvic organs in women, focal

FDG uptake was frequently seen in the normal uterine

endometrium and ovaries of premenopausal women in

certain phases of the menstrual (or ovarian hormonal)

Received: 18 April 2008 / Accepted: 13 June 2008 The Japanese Society of Nuclear Medicine 2008

Abstract

Objective Uterine leiomyomas sometimes show focal18F-fluorodeoxyglucose (FDG) uptake on positron

emission tomography (PET) images that may result in a

false-positive diagnosis for malignant lesions. This study

was conducted to investigate the incidence and charac-

teristics of uterine leiomyomas that showed FDG

uptake.

Methods We reviewed FDG-PET and pelvic magnetic

resonance (MR) images of 477 pre-menopausal (pre-

MP, age 42.1 7.3 years) and 880 post-MP (age 59.9

6.8 years) healthy women who underwent these tests as

parts of cancer screening. Of 1357, 323 underwent annual

cancer screening four times, 97 did three times, 191 did

twice, and the rest were screened once. Focal FDG

uptake (maximal standardized uptake value >3.0) in the

pelvis was localized and characterized on co-registered

PET/MR images.

Results Uterine leiomyomas were found in 164 pre-MP

and 338 post-MP women. FDG uptake was observed in

18 leiomyomas of 17 of the 164 (10.4%) pre-MP women

and in 4 leiomyomas of 4 of the 338 (1.2%) post-MP

women. The incidence was significantly higher in pre-

MP women than in post-MP women (chi-square, P


2/9

804 Ann Nucl Med (2008) 22:803810

13

cycle [46]. Several case reports also demonstrated that

uterine leiomyomas, although benign, show FDG uptake

on rare occasions that may cause false-positive diagnosis

for malignant lesions [710]. The objective of this study

was to investigate the incidence and characteristics of

uterine leiomyomas that showed FDG uptake from data

of a large number of healthy women who underwent

FDG-PET and pelvic magnetic resonance (MR) imaging

as parts of cancer screening.

Materials and methods

Subjects

We included a total of 1357 female subjects in this study,

477 premenopausal (pre-MP, age 42.1 7.3 years) and

880 post-menopausal (post-MP, age 59.9 6.8 years)

women, after excluding those who met the exclusion

criteria: (1) history and/or diagnosis of gynecological

malignancy or surgery, (2) receiving hormonal therapy,

and (3) blood sugar level over 150 mg/ml at the time of

PET examination. They underwent whole-body FDG-

PET and pelvic MR imaging as parts of cancer screening

in the Hamamatsu Medical Imaging Center. Medical

interviews, encompassing prior malignancy and gyneco-

logical surgery, menstrual status, and phase of the men-

strual cycle were conducted with all women. All women

underwent the cancer screening at least once between

August 2003 and December 2006. Of 1357, 323 under-

went the annual cancer screening four times, 97 did threetimes, and 191 were screened twice.

Diagnoses of uterine leiomyomas were made on the

basis of findings of MR imaging and results of follow-up

till the end of 2007. Women with findings suggestive of

malignant lesions were referred to local hospitals for

further examinations or periodical follow-ups to obtain

the final diagnosis. Some women with findings suggestive

of leiomyomas were also referred to local hospitals

depending on the size and characteristics on MR images

and symptoms. We checked the occurrence of cancer

including gynecological malignancy 1 year after the

cancer screening by sending a questionnaire to womenwho did not receive further examinations or follow-ups.

Written informed consents were obtained from all

women for the study, which was approved by the ethics

committee of the Hamamatsu Medical Photonics

Foundation.

PET imaging

Positron emission tomography imaging was performed

with a dedicated PET scanner (SHR-92000, Hamamatsu

Photonics, Hamamatsu, Japan). The scanner has a long

axial field of view of 685 mm, containing 12 rows of

detector blocks (60 detector blocks in each row), which

produced 336 transverse sections with a section thickness

of 3.2 mm covering from the upper thigh to the top of

the brain in two bed positions with an effective axial

field of view of 1075 mm [11]. Each detector block has a

flat panel position sensitive-photomultiplier (PS-PMT)

(R8400-00-M64, Hamamatsu Photonics) and a 16 8

bismuth germanate (BGO) crystal array with a crystal

size of 2.9 mm 6.3 mm 20 mm. All women fasted for

at least for 5 h prior to being administered an injection of

FDG. The serum glucose levels were measured just prior

to the injection. All women voided immediately prior to

the scan, which was started 60 min following the injec-

tion of 3 MBq/(kg body-weight) FDG. A lower part of

the body was scanned first to avoid the degradation of

image quality by the urinary activity in the bladder. The

acquisition time was 7 min for one bed position.

Whole-body computed tomography (CT) with low

radiation dose (120 kV, 10 mAs, 0.5 s/rotation, effective

radiation dose of less than 0.5 mSv) was also obtained

with an 8-slice CT scanner (LightSpeed Ultra, GE

Medical Systems, Milwaukee, WI, USA) with holding

breath in an expiration phase, which was used for atten-

uation correction of the PET images. The position

and shape of the body at the time of the CT scan were

reproduced in the PET scanner using the vacuum molded

immobilization mattress (BlueBag Vacuum Cushion,

Medical Intelligence, Schwabmunchen, Augsburg,

Germany) made for each woman, which had been provedto be a practical device for reproducing the position of

the body [12, 13]. The PET images were reconstructed

by means of a dynamic row-action maximum likelihood

algorithm [14]. Reformatted transaxial, sagittal, coronal,

and maximum intensity projection (MIP) images were

used for the interpretation.

MR imaging

Magnetic resonance imaging was performed with a

1.5-T MR scanner (EXCITE, GE Medical Systems).A T2-weighted fast spin-echo (FSE) sequence was used

for transaxial [repetition time (ms)/echo time (ms) =

4300/102, 320 224 matrix], transaxial fat-saturation

(3700/102, 256 192 matrix), and sagittal (2400/102,

320 224 matrix) images. Two signals were averaged.

Coronal T1-weighted FSE images (470570/7.58.5,

320 224 matrix, one or two signal averaged) were also

obtained. All images were acquired with a 3036 cm field

of view, a 45 mm section thickness, and a 1-mm inter-

section gap.


3/9

Ann Nucl Med (2008) 22:803810 805

13

Image analysis

The FDG-PET images were evaluated for focal FDG

uptake in the pelvis visually and with standardized

uptake values (SUVs). The uptake value was corrected

for the injected dose and the body weight to obtain

SUVs, and the maximal SUVs (SUVmax) of the foci were

recorded. A focal area with FDG uptake showing an

SUVmaxgreater than 3 was considered to be positive. The

MR images were used to localize the foci of increased

FDG uptake and to evaluate morphological abnormal-

ity of the lesions. Anatomical correlation of FDG-PET

images with MR images was performed on co-registered

PET/MR images. For this purpose, we referred to CT

images obtained with a low radiation dose for attenua-

tion correction, which could be superimposed closely on

PET images. Anatomical markers such as bony struc-

tures of the pelvis were used for manual co-registration

of CT and MR images. PET images were then co-

registered on MR images.

Leiomyomas with FDG uptake were classified into

three groups according to the level of signal intensity on

T2-weighted MR images as low, almost equal (iso), and

high compared with that of myometrium to see the rela-

tionship between tissue characteristics and FDG uptake.

These were also classified into three groups according to

the levels of FDG uptake: SUVmax from 3 to 5 as mild

(+), from 5 to 8 as moderate (++), and over 8 as high

(+++), which was correlated with the menstrual status

and/or the phases of menstrual cycle at the examination

in each individual: the menstrual flow phase (M) fromday-1 to day-7 of the cycle, the follicular and periovula-

tory phases (F) from day 8 to 2 days after the expected

day of ovulation, and the luteal phase (L) for the rest of

the cycle.

Results

No woman developed uterine sarcoma in this study

although four women were diagnosed and proved to

have endometrial carcinomas.

Uterine leiomyomas were seen on T2-weighted MRimages in 164 of the 477 pre-MP women and in 338 of

the 880 post-MP women (Table 1). Twenty-two leiomyo-

mas with FDG uptake were found in 21 women. Details

of characteristics and findings of the 22 leiomyomas are

shown in Table 2. Eighteen leiomyomas with FDG

uptake were seen in 17 of the 164 (10.4%) pre-MP women

with the SUVmaxof 5.3 2.9 (range 3.516.0) and 4 were

seen in 4 of the 338 post-MP women (1.2%) with the

SUVmax of 6.1 2.3 (range 3.78.0). The incidence of

leiomyomas with FDG uptake was significantly higher

in pre-MP women than in post-MP women (chi-square,

P


4/9

806 Ann Nucl Med (2008) 22:803810

13

Table2

DetailsofcharacteristicsandfindingsofuterineleiomyomaswithF

DGuptakein21subjects

Subject

Initialstudy

1yearaftertheinitia

lstudy

2yearsaftertheinitialstudy

3yearsafterthein

itialstudy

No.

Age

Size(cm)

FDGuptake

(SUVmax)

MS

MRI

Size(cm)

FDGuptake

(SUVmax)

MS

MRI

Size(cm)

FDGuptake

(SUVmax)

MS

MRI

Size(cm)

FDGuptake

(SUVmax)

MS

MRI

1

45

1.7

L

Low

1.7

5.2++

L

Low

2.7

4.2+

F

High

4.2

3.0+

M

High

2

51

2.4

4.0+

L

Iso

2.4

4.1+

L

Iso

2.2

M

Iso

1.5

PM

Iso

3

49

2.7

4.5+

L

Low

3.0

M

Low

2.8

5.5++

L

Low

2.8

3.5+

Irreg

Low

2.2

3.9+

L

Iso

1.9

M

Iso

2.0

3.8+

L

Iso

2.1

3.0+

Irreg

Iso

4

44

2.0

5.3++

L

Iso

1.5

3.7+

Irreg

Iso

1.3

3.0+

Irreg

Iso

1.3

Irreg

Iso

5

40

9.2

8.0+++

L

Iso

8.5

4.5+

M

Iso

9.8

9.0+++

L

Iso

9.5

F

Iso

6

49

5.6

3.5+

F

Low

5.7

3.0+

F

Low

5.9

3.8+

M

Low

5.7

M

Low

7

55

2.3

8.0+++

PM(

4)

Iso

2.2

10.0

+++

PM(

5)

Iso

2.0

9.5+++

PM(

6)

Iso

1.8

8.5+++

PM(

7)

Iso

8

43

2.2

7.0++

L

Iso

2.1

4.1+

L

Iso

2.2

F

Iso

2.2

3.4+

L

Low

9

53

1.6

4.5+

PM(

2)

High

1.6

PM(

3)

High

1.5

PM(

4)

High

1.5

PM(

5)

Iso

10

49

2.0

3.8+

L

Iso

2.2

Irreg

Iso

1.8

PM

Iso

11

51

3.8

3.5+

M

Iso

3.8

Irreg

Low

12

44

3.5

L

Low

4.2

4.1+

L

Low

13

47

2.6

4.0+

Irreg

Iso

2.0

PM

Low

14

48

4.6

16.0

+++

M

Low

Post-hysterec

tomy

Post-hystere

ctomy

15

53

3.4

4.3+

M

Iso

16

48

3.1

4.3+

M

High

17

31

1.3

4.2+

Irreg

Low

18

51

3.5

3.7+

PM(

1)

Iso

19

46

3.0

3.8+

Irreg

Low

20

55

2.6

8.0+++

PM(

2)

Low

21

41

2.4

4.5+

L

Low

Hormonaltherapywasstartedforleiomyomasinsubjectno.4aftertheinit

ialstudy

MSmenstrualstatusandphases,M

menstrualflowphase,Ffollicularandperiovulatoryphases,Llutealphase,PM

post-menopausewithaperiod(years)aftermenopausein

parentheses


5/9

Ann Nucl Med (2008) 22:803810 807

13

and a differential diagnosis with imaging tests is impor-

tant to avoid unnecessary surgery [16, 17]. The initial

report using FDG-PET for the differential diagnosis of

leiomyosarcomas from leiomyomas suggested that

uterine sarcomas which showed FDG uptake could beclearly differentiated from leiomyomas which did not

accumulate FDG [18]. However, several recent case

reports revealed that FDG uptake could be also seen in

benign uterine leiomyomas [710] and indicated that

FDG-PET could not be used for the differential diagno-

sis of leiomyosarcomas from leiomyomas.

In this article, for further understanding of uterine

leiomyomas with FDG uptake, we investigated the inci-

dence and characteristics of those leiomyomas from data

of 1357 healthy women who underwent FDG-PET and

pelvic MR imaging as parts of cancer screening. The

value and feasibility of cancer screening including FDG-PET for healthy individuals have not been tested and

clarified yet, and a prospective study is now underway

in our center to evaluate annual cancer screening includ-

ing FDG-PET in healthy volunteers [19]. Through the

interpretation of FDG-PET images of healthy women in

this large population, we have encountered many foci of

FDG uptake in the pelvis that should be regarded as

physiological variations and pitfalls [56]. Understand-

ing of these physiological and benign FDG uptakes in

such large populations of healthy subjects is of great

importance for correct interpretation of pathological

processes of FDG-PET images.

In this study, we found leiomyomas in about 35% of

both pre- and post-MP women with the prevalence com-

parable with those of published data [15]. Leiomyomaswith FDG uptake were much more common in pre-MP

women as compared with post-MP women. The inci-

dence of 10.4% of pre-MP women with leiomyomas was

significantly higher than that of 1.2% of post-MP women

with leiomyomas (chi-square, P


6/9

808 Ann Nucl Med (2008) 22:803810

13

regulated by several factors including expression of

glucose transporter-1 (GLUT-1) and hexokinase, the

number of viable tumor cells, microvessel density, tumor

cell proliferation, and the presence of inflammatory cells

[24], and the combination of factors involved in each

tumor may be different. In breast cancer, for example,the FDG uptake in the tumors was shown to be the

function of microvasculature, expression of GLUT-1

and hexokinase, number of tumor cells/volume, prolif-

eration rate, number of lymphocyte, and hypoxia-

inducible factor-1 for upregulating GLUT-1 [25].

There are few reports regarding factors that regulate

FDG uptake in leiomyomas. In a recent report of three

cases, histopathological analysis showed increased vas-

cularity as a common finding, but there was no associa-

tion between proliferative activity evaluated by Ki67 and

FDG uptake [9]. In our study, only one woman (subject

no. 14 inTable 2) with multiple leiomyomas underwent

surgery, and immunohistochemical analysis showed

positive for proliferating cell nuclear antigen but there

was no difference in positive indices among leiomyomas

with and without FDG uptake. In this case, signal inten-sity of the leiomyoma with FDG uptake on T2-weighed

MR images was higher than that of leiomyomas without

FDG uptake. The finding was seen in majority of cases

in our study and seemed to be one of the characteristics

of leiomyomas with FDG uptake.

The finding of increased signal intensity in leiomyo-

mas on T2-weighed MR images is known to suggest

cellular leiomyomas with dense cellular components

with little or no collagen [17]. Cellularity has been

reported as one of the factors that affect FDG uptake in

a b

dc

Fig.2 The MIP (left) andsagittal (upperright) imagesof FDG-PET and T2-weighted sagittal MR images(lowerright) of a 40-year-oldpre-MP woman whounderwent annualexaminations four times in

April 2004 (a), March 2005(b), February 2006 (c), andMarch 2007 (d) are shown.Intense FDG uptake was seenin the large leiomyoma whichshowed signal intensityalmost equal to that ofmyometrium on the initialexamination that was done inthe late luteal phase of themenstrual cycle (a). On thesecond examination done onthe third day of the menstrualflow phase, the leiomyomashowed slight FDG uptake

(b). Intense FDG uptake wasseen again on the thirdexamination done in the lateluteal phase (c). There was noFDG uptake on the fourthexamination done in theperiovulatory phase (d). Therewas no interval change in thesize of the leiomyoma,whereas a small leiomyomawith low signal intensitywithout FDG uptake (arrows)showed a slight increase inthe size (ad)


7/9


8/9

810 Ann Nucl Med (2008) 22:803810

13

19. Nishizawa S, Inubushi M, Okada H, Ozawa F, Kojima S,Teramukai S, et al. Cancer screening trial to evaluate theefficacy of FDG PET in healthy subjects: 2-year results of theHamamatsu Medical Imaging Center study (abstract). J CLinOncol 2006;24 Suppl 18:1025.

20. Maruo T, Ohara N, Wang JW, Matuo H. Sex steroidal regula-tion of uterine leiomyoma growth and apoptosis. HumanReprod Update 2004;10:20720.

21. Maruo T, Matsuo H, Samoto T, Shimomura Y, Kurachi O,Gao Z, et al. Effects of progesterone on uterine leiomyomagrowth and apoptosis. Steroid 2000;65:58592.

22. Pavlovich SV, Volkov NI, Burlev VA. Proliferative activityand level of steroid hormone receptors in the myometrium andmyoma nodes in different phases of menstrual cycle. Bull ExpBiol Med 2003;136:3968.

23. Kawaguchi K, Fujii S, Konishi I, Nanbu Y, Mori T. Mitoticactivity in uterine leiomyoma during the menstrual cycle. AmJ Obstet Gynecol 1989;160:63741.

24. Buck AK, Reske SN. Cellular origin and molecular mecha-nisms of 18F-FDG uptake: is there a contribution of the endo-thelium? J Nucl Med 2004;45:4612.

25. Bos R, von der Hoeven JJM, von der Wall E, von derGroep P, van Diest PJ, Comans EFI, et al. Biologic correlatesof 18F-FDG uptake in human breast cancer measured bypositron emission tomography. J Clin Oncol 2002;20:37987.

26. Ito K, Kato T, Ohta T, Tadikoro M, Yamada T, Ikeda M,et al. Fluorine-18 fluoro-2-deoxyglucose positron emissiontomography in recurrent rectal cancer: relation to tumour size

and cellularity. Eur J Nucl Med 1996;23:13727.27. Berger KL, Nicholson SA, Dehdashti F, Siegel BA. FDG PETevaluation of mucinous neoplasms: correlation of FDG uptakewith histopathologic features. Am J Roentgenol 2000;174:10058.

28. Lippitz B, Cremerius U, Mayfrank L, Bertalanffy H, RaoofiR, Weis J, et al. PET-study of intracranial meningiomas: cor-relation with histopathology, cellularity and proliferation rate.Acta Neurochir Suppl 1996;65:10811.

29. Higashi T, Tamaki N, Torizuka T, Nakamoto Y, SakaharaH, Kimura T, et al. FDG uptake, GLUT-1 glucose trans-porter and cellularity in human pancreatic tumors. J NuclMed 1998;39:172735.


9/9

Reproducedwithpermissionof thecopyrightowner. Further reproductionprohibitedwithoutpermission.

Journal Leiomyoma

Documents

Transcript of Journal Leiomyoma