Download Defining an Imaging Algorithm for Noncystic Splenic Lesions

Pe d i a t r i c I m a g i n g • O r i g i n a l R e s e a r c h
Dhyani et al.
Imaging Noncystic Splenic Lesions
Downloaded from www.ajronline.org by Countway Library Of Medicine on 11/25/13 from IP address 128.103.149.52. Copyright ARRS. For personal use only; all rights reserved
Pediatric Imaging
Original Research
Defining an Imaging Algorithm for
Noncystic Splenic Lesions Identified
in Young Patients
Manish Dhyani1
Sudha A. Anupindi2
Rama Ayyala3
Peter F. Hahn1
Michael S. Gee1
Dhyani M, Anupindi SA, Ayyala R, Hahn PF, Gee
MS
Keywords: algorithm, computed tomography,
malignancy, MRI, non-cystic, pediatric imaging, spleen,
splenic lesion, ultrasound, young adults
DOI:10.2214/AJR.12.10105
Received October 10, 2012; accepted after revision
March 5, 2013.
1
Department of Radiology, Massachusetts General
Hospital, Harvard Medical School, 55 Fruit St, Ellison 237,
Boston, MA 02114. Address correspondence to M. S. Gee
([email protected]).
2
Department of Radiology, Children’s Hospital of Philadelphia, University of Pennsylvania Perleman School of
Medicine, Philadelphia, PA.
3
Department of Radiology, Columbia Presbyterian
Medical Center, Columbia University School of Medicine,
New York, NY.
WEB
This is a web exclusive article.
AJR 2013; 201:W893–W899
0361–803X/13/2016–W893
© American Roentgen Ray Society
OBJECTIVE. The purpose of this study was to classify noncystic splenic lesions detected on imaging in young patients (0–30 years) and to determine the optimal imaging workup
for such lesions.
MATERIALS AND METHODS. This study was conducted at three academic institutions by performing a database search of radiology reports (2002–2011) to identify patients
with noncystic splenic lesions. Medical records were then searched to identify radiology examination indications, clinical follow-up, and lesion changes on subsequent imaging. All lesions had either definitive diagnosis (histopathology or laboratory results consistent with infectious cause) or lesion stability more than 2 years consistent with a benign cause.
RESULTS. Benign (n = 32), benign indeterminate (n = 7), and malignant (n = 14) lesions were identified in 53 patients (26 males and 27 females; mean age, 19 years; age range,
1 month–30 years). Lesions were initially detected on the following imaging modalities: CT
(n = 27), ultrasound (n = 12), MRI (n = 6), and PET/CT (n = 8). A total of 14 patients underwent MRI for lesion characterization, and 12 underwent PET/CT. MRI permitted definitive
characterization of benign lesions in 10 of 14 (72%) patients, whereas PET/CT was used to
diagnose nine of nine (100%) malignant splenic lesions and helped exclude malignancy in
two of three benign lesions.
CONCLUSION. Contrast-enhanced MRI is recommended for imaging workup of noncystic splenic lesions discovered in young patients because it can enable definitive diagnosis
of most benign lesions. Lesions with indeterminate MRI features can be followed-up with ultrasound or CT. PET or PET/CT is recommended for patients with clinical evidence of malignancy but is less helpful for characterization of isolated splenic lesions.
D
espite its important role in the immune system, the spleen is often
referred to as the forgotten organ
[1]. Solitary focal lesions of the
spleen are relatively rare and hence tend to be
overlooked on imaging examinations [2]. This
low incidence of focal abnormality within the
spleen may be attributed to the high concentration of splenic immune effector cells that provide a good defense against infectious organisms and tumor infiltration. Despite these
barriers, a diverse range of lesions involve the
spleen, including congenital, infectious, inflammatory, traumatic, vascular, and neoplastic causes [3]. Simple cysts of the spleen can be
easily diagnosed. However, despite advances
in imaging technology that have led to increased conspicuity of splenic lesions, it remains difficult to distinguish benign from malignant noncystic splenic lesions on the basis of
imaging characteristics.
Most splenic lesions detected are incidental findings on abdominal imaging examinations performed for nonspleen-related clinical
indications and are commonly referred to as
“incidentalomas” [1]. Such incidentally discovered splenic lesions are frequently diagnosed, particularly in the emergency department setting where CT of the abdomen and
pelvis is often performed to evaluate patients
with abdominal pain or a history of blunt abdominal trauma. Additionally, the rise in abdominal ultrasound use for imaging evaluation of symptomatic young patients will likely
contribute to this increase. Such incidentally
found lesions represent a clinical dilemma regarding management, and in the absence of
a conclusive radiologic diagnosis, histopathologic evaluation by either percutaneous splenic biopsy or splenectomy may be required [4].
The management of splenic lesions found
in the young patient population (0–30 years)
AJR:201, December 2013W893
Downloaded from www.ajronline.org by Countway Library Of Medicine on 11/25/13 from IP address 128.103.149.52. Copyright ARRS. For personal use only; all rights reserved
Dhyani et al.
A
B
Fig. 1—7-year-old girl with echinococcal cysts
in spleen. Sagittal ultrasound image shows
cystic structure with distinct internal undulating
membranes. Red and blue pixels signify areas of
minimal internal Doppler vascularity.
Fig. 2—11-year-old girl with diffuse lymphangiomatosis.
A, Coronal T1-weighted image shows multiple lesions low in signal intensity in enlarged spleen (white arrows).
In addition patient has ascites (black arrow).
B, Coronal T2-weighted image shows bright lesions in spleen (white arrows) and bones (black arrow).
is a particular diagnostic dilemma because
of the relatively low rates of invasive cancer
and cancer-related death in young patients
compared with the older patient population
[5]. The decision whether to seek a histologic diagnosis for a splenic incidentaloma
that is most likely benign is difficult given
the low, but not zero, complication and morbidity rates associated with splenic biopsy
[6, 7] or splenectomy [8]. This highlights the
need to investigate the role of different imaging modalities in characterization of noncystic splenic lesions. The purpose of our study
was therefore twofold. The first goal was to
determine the causes of incidentally discovered splenic lesions in young patients on the
basis of a relatively large number of patients
from three academic medical centers that see
a high volume of pediatric and young adult
patients. The second goal was to derive a
practical imaging algorithm for evaluation of
incidentally discovered splenic lesions in the
0- to 30-year-old age group.
matic splenic lacerations were excluded. Nonsimple cysts were included because they can arise from
infection or necrotic tumor. In addition, lesions that
did not have a histologic diagnosis or meet the criteria for diagnosis as a “benign indeterminate” lesion (defined later) were excluded. Fifty-three patients were identified within the age group of 0–30
years with focal splenic lesions detected using ultrasound, CT, MRI, or PET.
We reviewed the medical records of all 53 patients. The following were noted in a detailed
data acquisition chart: initial mode of scanning at
which splenic lesions were first seen, age at initial scanning, reason for imaging, and subsequent
follow-up. Medical and surgical records were then
reviewed along with histology and laboratory results. The final diagnosis of each lesion was re-
corded on the basis of the histopathology report
from either splenic biopsy or surgical splenectomy. Lesions without a definitive histopathologic
diagnosis were included in the study and assigned
a diagnosis of “benign indeterminate” if they
showed stable appearance on subsequent imaging
examinations performed at least 2 years after the
date of the initial imaging examination that documented the presence of the lesion.
The reasons for initial scanning of the patient
were grouped under the following categories: trauma
workup, when imaging was performed in the emergent setting as part of the trauma workup; infectious
cause, if medical or surgical notes documented clinical signs of an infection after which imaging was
performed to search for an infectious source; abdominal condition, all lesions that were detected on
imaging performed for symptoms and signs of abdominal pain or hepatosplenomegaly and workup
A
B
Data Collection
Materials and Methods
After institutional review board approval, which
waived informed patient or parent consent, a retrospective review of radiology reports was performed at three tertiary care centers (one dedicated
pediatric hospital and two large tertiary care centers). Radiology databases were searched for keywords, such as “splenic lesion,” “splenic mass,”
and “splenic focus,” for a period between 2002 and
2011. Only patients under the age of 30 years at the
time of detection of the lesion were included in the
study. Lesions consistent with simple cysts or trau-
W894
Fig. 3—5-year-old girl with surgically proven splenic lymphangioma.
A and B, Contrast-enhanced axial (A) and early delayed (B) CT images of focal complex low-attenuation mass
(arrow) show minimal enhancement,
AJR:201, December 2013
Downloaded from www.ajronline.org by Countway Library Of Medicine on 11/25/13 from IP address 128.103.149.52. Copyright ARRS. For personal use only; all rights reserved
Imaging Noncystic Splenic Lesions
for abnormal laboratory values (elevated liver function tests, neutropenia, positive d-dimer); lymphangiomatosis, three patients with lymphangiomatosis
were grouped under this category; evaluation of malignant disease, lesions detected on staging examinations performed for evaluation of malignant disease
were categorized in this group.
Analysis
A
B
Fig. 4—11-year-old boy who presented with splenomegaly.
A, Contrast-enhanced coronal HASTE T2-weighted image without fat saturation shows marked enlargement of
spleen with multiple high-signal-intensity lesions within spleen (arrows).
B, Contrast-enhanced coronal gradient-echo image shows flow voids in some lesions (arrows), consistent with
multiple hemangiomas.
A
B
Fig. 5—24-year-old man with lesion in spleen.
A and B, Coronal (A) and axial (B) contrast-enhanced MR images show large lesion in spleen (white arrow)
and smaller lesions in liver (black arrows), consistent with metastases (bronchoalveolar epithelioid
hemangioendothelioma).
Fig. 6—30-year-old
woman with breast
angiosarcoma. Axial
contrast-enhanced
CT image shows right
hepatic liver lobe
with heterogeneous
mass (arrow) and
smaller lesion in
spleen, consistent with
metastases.
All lesions were categorized as benign, benign
indeterminate, or malignant. Benign and malignant lesions were categorized on the basis of histopathology and definitive diagnosis on MRI with
IV contrast administration or PET. For benign lesions, histopathology, if acquired, was considered
diagnostic and imaging features diagnostic of a
specific benign condition were reassuring, with
a minimum of 6 months of follow-up qualifying
as a benign categorization. Benign indeterminate
lesions were categorized on the basis of stability
shown by serial follow-up imaging for 2 years. Lesions that were stable but were not characterized
on the basis of imaging features and lacked 2-year
follow-up were excluded from the study. All malignant lesions had biopsy-proven histopathology
of the primary lesion, and imaging follow-up of
the splenic lesion confirmed the characteristics of
a malignant cause.
All images were reviewed by three of the authors, who have more than 5 years of experience
in pediatric abdominal imaging. The imaging examinations performed for the patients were associated with the final diagnoses to assess the utility
of each of the imaging modalities.
Results
There were a total of 53 patients (26 males
and 27 females) with a mean age of 19 years
(age range, 1 month–30 years) identified
with focal splenic lesions. A total of 39 benign lesions were identified (16 males and
23 females; mean age, 18 years; age range, 1
month–30 years) (Figs. 1–4). We had a total
of 14 malignant lesions (10 males and four
females; mean age, 24 years; age range, 17–
30 years) (Figs. 5 and 6). All splenic lesions
that were evaluated under the three categories; benign (n = 32), benign indeterminate
(n = 7), and malignant (n = 14) are listed
in Table 1. The number of imaging examinations performed for the diagnosis of each
type of lesion were evaluated: benign (22
ultrasound, 33 CT, 18 MRI, one PET), benign indeterminate (6 ultrasound, 16 CT, one
MRI, four PET), and malignant (17 CT, two
MRI, and 14 PET).
The initial modality on which each lesion
was seen and the reason the imaging was per-
AJR:201, December 2013W895
Dhyani et al.
Downloaded from www.ajronline.org by Countway Library Of Medicine on 11/25/13 from IP address 128.103.149.52. Copyright ARRS. For personal use only; all rights reserved
TABLE 1: Splenic Lesions and Diagnostic Workup
Diagnostic Workup
Patient
No.
Lesions
Ultrasound
CT
MRI
PET
1
2
1
Benign
1
SANT
1
2
Littoral cell
—
1
1
—
3
Lymphangiomatosis
1
—
1
—
4
Lymphangiomatosis
1
—
1
—
5
Lymphangiomatosis
1
—
1
—
6
Lymphangioma
1
1
—
—
7
Hamartoma
1
1
—
—
8
Hemangioma
1
1
1
—
9
Hemangioma
—
1
1
—
10
Hemangioma
—
1
—
—
11
Hemangioma
—
1
1
—
12
Hemangioma
1
—
—
—
13
Complex cyst
2
1
—
—
14
Complex cyst
3
1
—
—
15
Complex cyst
2
—
1
—
16
Complex cyst
1
2
1
—
17
Proteinaceous cyst
1
—
1
—
18
Pseudocyst
1
1
—
—
19
Infarcts
—
2
—
—
20
Infarcts
—
1
1
—
21
Infarcts
—
1
—
—
22
Infarcts
—
2
1
—
23
Infarcts
—
1
—
—
24
Bartonella
1
1
1
—
25
Bartonella
—
1
1
—
26
Bartonella
—
1
—
—
27
Echinococcus
—
2
—
—
28
Echinococcus
1
1
—
—
29
Mycobacterium tuberculosis
—
3
—
—
30
Candida
1
1
—
—
31
Splenic abscess
—
3
1
—
32
Siderotic nodule
1
—
1
—
6
16
1
4
33–39
Benign Indeterminate
Malignant
40–48
Lymphoma
—
8
1
14
49
Sarcoma
—
2
—
—
50
Angiosarcoma (breast)
—
1
—
—
51
Epithelioid hemangioendothelioma
—
1
1
—
52
Teratoma (testis)
—
3
—
—
53
Seminoma (testis)
—
2
—
—
28
66
21
19
Total
Note—SANT = sclerosing angiomatoid nodular transformation. Dash indicates not applicable.
W896
formed are tabulated in Table 2. Excluding
lesions found in patients undergoing followup of malignancy, splenic lesions were most
commonly seen when patients were evaluated for trauma (14/38, 36.8%), with a majority
(13/14) benign or benign indeterminate. For
benign and benign indeterminate lesions, incidental splenic lesions were most commonly seen when the patient was being evaluated
for trauma (13/39, 33%) and abdominal pain
(13/39, 33%). Malignant lesions of the spleen
were most commonly seen during evaluation
of a known malignancy (13/14, 93%), with
only one of 14 (7%) diagnosed in a patient
without a known history of malignancy.
For 13 patients being evaluated for abdominal pain, ultrasound was the initial imaging modality in six of 13, five of 13 with
CT, and two of 13 with MRI. CT was the
most common initial imaging modality used
for evaluating patients for trauma workup
(11/13) and infection (6/8). MRI was the first
imaging modality for patients with an abnormal physical examination (2/3), whereas PET/CT was the initial imaging modality
used in evaluating patients with known malignant disease (8/15).
Most lesions identified on imaging performed for trauma patients (13/14) were benign (93%), except in one patient in whom
an incidental finding on CT led to the diagnosis of diffuse large B cell lymphoma. On
the other hand, only two of 15 (13%) lesions
on follow-up imaging for known malignant
disease were benign. For imaging performed
in patients without a known history of malignancy (n = 38) (i.e., imaging performed for
trauma workup, infectious cause, abdominal
condition, and lymphangiomatosis), only one
of 38 (2.6%) resulted in a malignant finding.
The basis of diagnosis of all benign and
benign indeterminate lesions is tabulated in
Table 3. MRI was performed in 14 of 39 patients and was diagnostic in 10 (71%). Of the
four patients (sclerosing angiomatoid nodular transformation [SANT], littoral cell angioma, Bartonella infection, and benign indeterminate) with nondiagnostic MRI, two
underwent subsequent PET (SANT and Bartonella infection). For the patient with Bartonella infection, the diagnosis was established by positive serology correlation and
response to treatment on subsequent imaging for a definitive diagnosis. The littoral cell
angioma was nondiagnostic on biopsy and
the patient underwent splenectomy for definitive diagnosis. Overall, PET was performed
for three benign lesions (8%), and a negative
AJR:201, December 2013
Imaging Noncystic Splenic Lesions
TABLE 2: Reason for Imaging, Initial Imaging Modality Used, and Final Diagnosis
Downloaded from www.ajronline.org by Countway Library Of Medicine on 11/25/13 from IP address 128.103.149.52. Copyright ARRS. For personal use only; all rights reserved
Initial Modality
Reason for Imaging
Ultrasound
CT
Final Diagnosis
MRI
PET
Benign
Benign Indeterminate
Malignant
Total
Trauma workup
2
11
1
—
11
2
1
14
Infectious cause
2
6
—
—
8
—
—
8
Abdominal condition
6
5
2
—
10
3
—
13
Lymphangiomatosis
1
—
2
—
3
—
—
3
Evaluation of malignant disease
1
5
1
8
—
2
13
15
Total
12
27
6
8
32
7
14
53
Note—Dash indicates not applicable.
TABLE 3: MRI and PET for Benign and Benign Indeterminate Lesions
MRI
Final Diagnosis
No.
PET
No.
Diagnosis
No.
FDG Uptake
Basis of Diagnosis
Infectious
Bartonella
3
1
Not conclusive
—
—
2 Histopathology (liver biopsy) and 1 treatment response
Echinococcus
2
—
—
—
—
Imaging and clinical correlation
Candida
1
—
—
—
—
Clinical correlation
Mycobacterium tuberculosis
1
—
—
—
—
Histopathology (autopsy)
Abscess
1
—
—
—
—
Clinical correlation
Benign indeterminate
7
1
Not conclusive
2
No
Follow-up imaging
Hemangioma
5
3
Conclusive
—
—
3 MRI, histopathology (autopsy), histopathology (splenectomy)
Pseudocyst
1
—
—
—
—
CT
Lymphangiomatosis
3
3
Conclusive
—
—
MRI
Proteinaceous cyst
1
1
Conclusive
—
—
MRI
Siderotic nodule
1
1
Conclusive
—
—
MRI
Infarct
5
1
Conclusive
—
—
4 CT, 1 MRI
Complex cyst
4
1
Conclusive
—
—
2 Histopathology (biopsy), 1 MRI, 1 histopathology (splenectomy)
Littoral cell angioma
1
1
Not conclusive
—
—
Histopathology (splenectomy)
Hamartoma
1
—
—
—
—
Histopathology (splenectomy)
Lymphangioma
1
—
—
—
—
Histopathology (splenectomy)
SANT
1
1
Not conclusive
1
Yes
Histopathology (splenectomy)
Note—SANT = sclerosing angiomatoid nodular transformation. Dash indicates not applicable.
TABLE 4: MRI and PET for Diagnosis of Malignant Lesions
MRI
Final Diagnosis
No.
No.
Nodular sclerosing type
7
Large B-cell
2
Sarcoma
PET
Diagnosis
No.
FDG
Basis of Diagnosis
—
—
7
Uptake
Clinical and PET
1
Conclusive
2
Uptake
Clinical and PET
1
—
—
—
—
Clinical
Angiosarcoma (breast)
1
—
—
—
—
Clinical
Epithelioid hemangioendothelioma
1
1
Conclusive
—
—
Clinical
Teratoma (testis)
1
—
—
—
—
Clinical
Seminoma (testis)
1
—
—
—
—
Clinical
Lymphoma
Metastases
Note—Dash indicates not applicable.
AJR:201, December 2013W897
Dhyani et al.
Downloaded from www.ajronline.org by Countway Library Of Medicine on 11/25/13 from IP address 128.103.149.52. Copyright ARRS. For personal use only; all rights reserved
Noncystic splenic lesion
No known malignancy
History of malignancy
Not FDG avid
MRI
Definitely benign
(Hemangioma, lymphangioma, etc.)
PET or PET/CT
Indeterminate MRI
FDG avid
Follow-up imaging
(Ultrasound or CT)
(Ultrasound preferred if lesion is visible)
Consider
histological
sampling
Workup for
malignancy
Fig. 7—Flowchart shows proposed algorithm for follow-up of splenic lesions.
PET provided reassurance against malignant
disease in two (67%) patients. Both patients
were subsequently followed up for more than
2 years with stability of lesion and were categorized as benign indeterminate. A positive
PET result for one lesion resulted in subsequent splenectomy and diagnosis of SANT
on histopathology. Negative PET results and
stability of splenic lesion for more than 2
years helped exclude malignant disease in
two patients being evaluated for Hodgkin
lymphoma and gallbladder carcinoma. Histopathology without MRI, was acquired by
biopsy (n = 2) and surgery (n = 4) in a total
of six noninfectious benign lesions.
Table 4 summarizes the utility of MRI
and PET in the determination of malignant
splenic lesions. PET was performed in 10 of
14 patients and showed 18F-FDG-avidity in
100% of patients.
Discussion
To our knowledge, an imaging algorithm
for incidentally discovered splenic lesions in
young patients has not been established. This
is in part because no comprehensive review of
splenic lesions in young patients with a known
diagnosis has yet been performed. We suspect
that most splenic lesions detected on imaging
would be benign, given the lower baseline incidence of overall malignancy in young people relative to older adults [9]. Our study of
three large academic centers, including one
dedicated pediatric hospital and two large tertiary care referral centers, revealed a total of
53 splenic lesions detected in young patients
on imaging, with 39 benign and 14 malignant lesions for a 74% rate of benignity. Thirteen of the 14 malignant lesions were identi-
W898
fied on imaging examinations of patients with
a known history of extrasplenic malignancy.
When these patients are excluded, the rate of
benign lesions rises to 97% (38/39) in young
patients. To our knowledge, this represents
the largest report of imaged splenic lesions in
young patients to date.
Having established the reason for imaging and imaging modality that led to detection of the splenic lesions in our patients, we
then examined the follow-up imaging performed on each lesion to determine the utility of different imaging modalities on the
basis of their ability to aid the radiologist in
characterizing splenic lesions and offering a
definitive diagnosis. One unique feature of
young patients relative to older adults is that
ultrasound is used more frequently as the
primary imaging modality because of concerns about the ionizing radiation exposure
associated with CT [10–12]. In our series,
ultrasound was the most common initial imaging modality for detecting splenic lesions
in patients imaged for abdominal conditions
(6/13, 46%). However, CT was the primary
initial imaging modality in patients with a
history of trauma, infection, or malignancy
(22/37, 59%). CT and ultrasound were both
useful for the detection of splenic lesions and
for distinguishing cystic from noncystic lesions. However, in our series neither modality was considered sufficient to yield a definitive diagnosis of a noncystic splenic lesion on
the basis of the initial imaging. In the case of
CT, the initial diagnostic examinations performed were single portal venous phase examinations used for routine abdominal evaluation and lacked the arterial and delayed
phase images that would be more sensitive
for the diagnosis of certain benign splenic lesions, particularly proteinaceous cysts
and hemangiomas. In the case of ultrasound,
certain sonographic features were suggestive of benign proteinaceous cysts, echinococcal disease (Fig. 1), or hemangioma (hyperechogenicity, increased posterior acoustic
enhancement, internal septations), but follow-up imaging was still recommended to
increase diagnostic certainty.
MRI proved to be a useful imaging modality for the diagnosis of benign splenic lesions (i.e., hemangioma, siderotic nodules,
lymphangiomas, proteinaceous or hemorrhagic cysts, and infarcts) (Figs. 2–4). MRI
was performed in 14 benign splenic lesions
and was diagnostic in 10 (71%). The combination of T2-weighted, gradient-echo, and
multiphase contrast-enhanced imaging provides superior lesion characterization compared with CT or ultrasound. We believe that
MRI should be performed for all incidentally
detected splenic lesions because of its ability
to aid the radiologist in making a definitive
diagnosis for a number of benign entities.
PET was a helpful imaging modality for
evaluating splenic lesions in patients with a
history of malignancy, aiding a 100% accurate evaluation of splenic lesions in this population. The majority of malignant splenic
lesions (9/14, 64%) were foci of lymphoma
in patients with a history of lymphoma. In
contrast, the incidence of splenic metastases was very low in young patients without
a history of malignancy (1/38, 2.6%). The
current literature for adults estimates the incidence of nonlymphomatous splenic metastases at 1%, with the most common primary
sites being lung (21%), stomach (16%), pancreas (12%), liver (9%), and colon (9%) [3].
In our study population, PET was performed
for three splenic lesions in patients without
a history of malignancy. All three lesions
were benign, with one lesion showing FDG
avidity on PET that led to splenectomy and
the diagnosis of benign SANT. PET helped
classify one lesion as benign in a patient with
known Hodgkin lymphoma, resulting in a
7% (1/15) rate of benign disease in patients
with a known history of malignancy. Our results suggest that PET is indicated for splenic lesion evaluation in patients with a history
of malignancy because of the relatively high
risk of metastases. In contrast, in patients
with no history of malignancy, our data do
not support routine PET evaluation of splenic
lesions due to the low baseline risk of splenic
metastases and the potential for false-posi-
AJR:201, December 2013
Downloaded from www.ajronline.org by Countway Library Of Medicine on 11/25/13 from IP address 128.103.149.52. Copyright ARRS. For personal use only; all rights reserved
Imaging Noncystic Splenic Lesions
tive FDG uptake by benign splenic lesions,
including hamartoma [13] and SANT [14].
A potential imaging algorithm (Fig. 7) for
young patients with incidentally discovered
splenic lesions includes MRI with IV contrast administration, which could provide definitive imaging diagnosis of some benign lesions. For lesions that are not clearly benign,
the next step will vary depending on whether the patient has a known history of malignancy. In patients with a history of malignancy, PET or PET/CT can be helpful to exclude
FDG avidity suggestive of malignancy. In
young patients with no history of malignancy,
an MRI-indeterminate lesion is most likely
benign. Management options we suggest from
our experience include serial imaging and biopsy (particularly if the patient is symptomatic). Serial imaging should be performed using
ultrasound if a lesion is readily distinguished
and measured sonographically; otherwise,
MR or CT should be considered.
There are limitations of our study. This
was a retrospective review of medical records
and some patients may have been missed. In
addition, only patients who were referred for
imaging examinations were included in our
cohort, and they may not be truly representative of the young population at large. However, this is the population that forms the
basis of our imaging algorithm. Also, histopathology was not available for some of the
benign lesions in our study; however, diagnosis by characteristic imaging features or
stability over time is widely accepted as a
surrogate for histology for these lesions. The
small number of patients in our study, albeit
relatively large compared with those in the
reported literature, precludes rigorous statistical analysis and limits our evaluation
of the results. Finally, all of the institutions
that participated in this study are located in a
similar geographic area (northeastern United
States); the endemic population of this area
may have its own bias with respect to infectious and demographic variation.
Conclusion
Splenic lesions are uncommon and pose a
dilemma for the radiologist and clinician. To
our knowledge, this is the largest imaging presentation of splenic lesions in the 0–30-yearold population. Our data show that most of
these noncystic splenic lesions are benign, and
in many cases contrast-enhanced MRI can provide reassurance. We provide a potential imaging algorithm to demystify these problematic
lesions in the “forgotten organ” and provide a
path of management to reduce the need for invasive procedures in young patients.
References
1.Ahmed S, Horton KM, Fishman EK. Splenic incidentalomas. Radiol Clin North Am 2011; 49:323–347
2.Warshauer DM, Hall HL. Solitary splenic lesions.
Semin Ultrasound CT MR 2006; 27:370–388
3.Sutherland T, Temple F, Hennessy O, Lee WK.
Abdomen’s forgotten organ: sonography and CT
of focal splenic lesions. J Med Imaging Radiat
Oncol 2010; 54:120–128
4.Makrin V, Avital S, White I, Sagie B, Szold A.
Laparoscopic splenectomy for solitary splenic tumors. Surg Endosc 2008; 22:2009–2012
5.Siegel R, Desantis C, Virgo K, et al. Cancer treat-
ment and survivorship statistics, 2012. CA Cancer
J Clin 2012; 62:220–241
6.Lieberman S, Libson E, Sella T, Lebensart P,
Sosna J. Percutaneous image-guided splenic procedures: update on indications, technique, complications, and outcomes. Semin Ultrasound CT
MR 2007; 28:57–63
7.Keogan MT, Freed KS, Paulson EK, Nelson RC,
Dodd LG. Imaging-guided percutaneous biopsy
of focal splenic lesions: update on safety and effectiveness. AJR 1999; 172:933–937
8.Jugenburg M, Haddock G, Freedman MH, FordJones L, Ein SH. The morbidity and mortality of
pediatric splenectomy: does prophylaxis make a
difference? J Pediatr Surg 1999; 34:1064–1067
9.Howlader N, Noone AM, Krapcho M, et al., eds.
SEER cancer statistics review, 1975–2009 (vintage 2009 populations), National Cancer Institute
website. seer.cancer.gov/csr/1975_2009_pops09/.
Published November 2011. Accessed July 25, 2013
10.Brenner D, Elliston C, Hall E, Berdon W. Estimated risks of radiation-induced fatal cancer from
pediatric CT. AJR 2001; 176:289–296
11.Faulkner K, Moores BM. Radiation dose and somatic risk from computed tomography. Acta Radiol 1987; 28:483–488
12.Sankaranarayanan K. Estimation of the genetic
risks of exposure to ionizing radiation in humans:
current status and emerging perspectives. J Radiat Res 2006; 47(suppl B):B57–B66
13.Avila L, Sivaprakasam P, Viero S, et al. Splenic
hamartoma in a child in the era of PET-CT. Pediatr Blood Cancer 2009; 53:114–116
14.Bamboat ZM, Masiakos PT. Sclerosing angiomatoid nodular transformation of the spleen in an
adolescent with chronic abdominal pain. J Pediatr Surg 2010; 45:E13–E16
AJR:201, December 2013W899