Downloaded from www.ajronline.org by 103.47.135.13 on 06/20/16 from IP address 103.47.135.13. Copyright ARRS.

For personal use only; all rights reserved

Wooden Foreign Bodies:

Imaging Appearance
Jeffrey J. Peterson 1 OBJECTIVE. The purpose of this study was to identify the characteristic imaging features
Laura W. Bancroft of wooden foreign bodies.
Mark J. Kransdorf MATERIALS AND METHODS. The imaging studies of 12 patients with surgically
confirmed wooden foreign bodies were retrospectively reviewed. The study group consisted
of seven females and five males, 1065 years old (mean age, 36 years). All patients underwent
radiography. Nine patients were evaluated with sonography, eight with MR imaging, three
with CT, and one with CT arthrography. Gadolinium-enhanced MR imaging was performed
in six patients. Three patients presented with a draining sinus and nine with painful swelling.
Only three patients presented with a history of penetrating injury.
RESULTS. Lesions were located in the foot (n = 4), hand (n = 3), thigh (n = 2), calf (n =
2), and elbow (n = 1). Radiographs failed to reveal the retained foreign bodies in all patients.
With MR imaging, wooden foreign bodies displayed a variable signal intensity that was equal
to or less than that of skeletal muscle on both T1- and T2-weighted images. MR imaging
showed the surrounding inflammatory response in all patients. CT showed the retained wood
as linear cylindric foci of increased attenuation. Wood was highly echogenic and revealed
pronounced acoustic shadowing on sonography. Arthrography in one patient showed an asso-
ciated reactive synovitis.
CONCLUSION. The imaging appearance of wooden foreign bodies is variable; how-
ever, imaging can be quite specific, and when taken in the appropriate clinical setting, the im-
aging should reliably suggest the diagnosis. Sonography is frequently underused but proved
most useful for the evaluation of retained wooden foreign bodies.

Received May 21, 2001; accepted after revision
September 17, 2001.
Presented at the annual meeting of the American
Roentgen Ray Society, Seattle, AprilMay 2001.
1
All authors: Department of Radiology, Mayo Clinic,
4500 San Pablo Rd., Jacksonville, FL 32224-3899.
Address correspondence to M. J. Kransdorf.
AJR 2002;178:557562
0361803X/02/1783557
American Roentgen Ray Society
D
espite advances in imaging tech-
niques, the detection of retained
wooden foreign bodies remains a
difficult and challenging task. Patients often
present for evaluation several months or even
years after the initial injury, and consequently,
clinical evaluation may fail to elicit a history of
antecedent skin puncture. When a history of
penetrating trauma is suggested, its severity is
difficult to estimate clinically. Even when there
is a high suspicion of a retained foreign body,
localization remains difficult. Foreign body
fragments may remain in the wound even after
apparent successful extraction by the patient at
the time of injury. The initial physical examina-
tion may reveal a painful swollen soft-tissue
mass or psuedotumor that may simulate malig-
nancy or infection, rather than suggesting a re-
tained foreign body. The detection of wood is
especially important because it may serve as an
unrecognized nidus for infection. Wood, with its
porous consistency and organic nature, is an ex-
cellent medium for microorganisms, and the re-
tained wooden foreign matter may result in
cellulitis, abscess, or fistula formation [1]. The
wooden foreign matter may also result in syno-
vitis if a joint is violated or in osteomyelitis if
adjacent osseous structures become involved.
Radiography, CT, MR imaging, and sonography
have been advocated for the detection of retained
foreign bodies. We reviewed our experience
with 12 patients who had retained wooden for-
eign bodies, only three of whom presented with
a history of previous penetrating trauma.
Materials and Methods
We retrospectively reviewed the imaging studies
of 12 patients with surgically confirmed retained
wooden foreign bodies. The study group was com-
posed of seven female and five male patients, 1065
years old (mean age, 36 years). Radiographs were

AJR:178, March 2002 557

 Peterson et al.

available for each patient. Additional imaging in-
cluded the following: sonography (n = 9), MR imag-
ing (n = 8), CT (n = 3), and CT arthrography (n = 1).
Sonography was performed using high-resolution
phased array probes. MR imaging included spin-echo
T1-weighted sequences and either conventional spin-
echo dual-echo or short tau inversion recovery se-
Downloaded from www.ajronline.org by 103.47.135.13 on 06/20/16 from IP address 103.47.135.13. Copyright ARRS. For personal use only; all rights reserved
hand in three, the thigh in two, the calf in two, and Two patients revealed the retained fragment
the elbow in one. as a signal void (Fig. 2). The surrounding in-
flammatory response was seen in all patients.
The response was hypointense on T1-weighted
Results
Radiography images and isointense to hyperintense on T2-
weighted images in relation to surrounding
Radiographs failed to reveal the retained

quences. Gadolinium-enhanced imaging was available
in six patients. All CT scanning was performed in the
axial plane with either a 1- or 3-mm slice thickness.
Reformatted images were obtained in the sagittal and
coronal planes. All CT scans were examined in both
soft-tissue and bone windows.
Patient presentation was variable. The most com-
mon presenting symptoms were pain and swelling,
which were observed in 10 patients. Three patients
presented with a draining sinus. Only three patients
presented prospectively with a history of penetrating
trauma. The foot was involved in four patients, the
foreign bodies in all patients. Bone erosion
was seen in one patient related to the close
proximity of the retained wood to the hu-
merus. The erosion was subtle, and although
identified prospectively, the adjacent foreign
body was not identified (Fig. 1).
MR Imaging
MR imaging showed the retained foreign
bodies to be hypointense to skeletal muscle
on both T1- and T2-weighted sequences.
skeletal muscle and fat, respectively (Figs.
24). Gadolinium-enhanced imaging in six
patients showed enhancement of the sur-
rounding inflammatory response.
CT
CT depicted the retained wooden foreign
body as a linear or cylindric area of high atten-
uation compared with the surrounding skeletal
muscle and fat (Fig. 5). All foreign bodies
were best visualized on images viewed at a

A B

Fig. 1.Retained wooden foreign body in 11-year-old boy with 2-year history of elbow pain and
swelling.
A, Anteroposterior elbow radiograph shows subtle extrinsic erosion (arrow) related to close
proximity of retained wooden foreign body.
B, Anteroposterior elbow radiograph after performance of arthrogram shows marked reactive
synovitis.
C, Axial CT scan of elbow after performance of arthrogram reveals subtle hyperattenuating
structure, which at surgery proved to be retained wooden foreign body (open arrow) with as-
sociated extrinsic erosion to posterior cortex of humerus (solid arrow).
C

558 AJR:178, March 2002

 Wooden Foreign Bodies

wide (bone) window. The surrounding inflam-

matory response seen on MR imaging was dif-
ficult to differentiate from surrounding skeletal
muscle, although effacement of surrounding
fat planes, indicative of an inflammatory re-
sponse, was seen in two patients.
Downloaded from www.ajronline.org by 103.47.135.13 on 06/20/16 from IP address 103.47.135.13. Copyright ARRS. For personal use only; all rights reserved

Sonography
Sonography showed the retained wooden for-
eign bodies as linear echogenic structures with
pronounced acoustic shadowing (Figs. 4 and 5).
In two patients with larger pieces of wood, only
the hyperechoic, crescent-shaped leading edge
of the wood was seen with acoustic shadowing
obscuring the trailing edge.

CT Arthrography
In one case in which the wooden foreign A
body was suspected to have traversed the el-
bow joint, athrography revealed an irregular
margin of the joint with a thickened, corru-
gated synovium (Fig. 1). Contrast material
was seen about the retained foreign body.

Discussion
Radiographs are usually unrewarding in
the search for retained foreign bodies, as
they were in this series. Radiographs have
been reported to reveal a wooden foreign
body in only 15% of patients [2]. Wooden
foreign bodies are usually radiolucent, asso-
ciated with gas in the matrix. However, the
small size of the foreign body often is not
sufficient to create an appreciable radiolu-
cency [3]. In one patient in our series, the
wooden foreign body came to rest in close
proximity to the distal humerus, with subse-
quent nonspecific extrinsic erosion of the un-
derlying cortex (Fig. 1). When retained
foreign bodies penetrate or lie adjacent to
bone, osteolytic, osteoblastic, or a combina-
tion of changes can occur [4]. Unfortunately,
this can often confuse the situation further
and suggest another underlying process other
than a retained splinter.
Xeroradiography has been reported as
slightly more sensitive than conventional ra-
diography for the detection of retained wood;
however, xeroradiographs show negative re-
sults in 80% of patients and are not available
in most radiology departments [2].
CT has been shown to be useful in the evalu-
ation for suspected wooden matter. In our se-
ries, retained wooden foreign bodies were
more subtle when using the standard window
and level setting. When the settings were al-
tered by increasing the window width, the for-
B
Fig. 2.49 year-old woman who stepped on toothpick 7 days before imaging.
A, Spin-echo T1-weighted axial MR image (TR/TE, 700/20) of forefoot shows retained wooden foreign body as fo-
cal signal void (open arrow) with surrounding hypointense inflammatory reaction (solid arrow).
B, Corresponding spin-echo T2-weighted axial MR image (2,000/90) depicts retained wood as signal void (open
arrow) with hyperintense surrounding foreign body granulomatous response (solid arrow).
eign bodies were more easily identified. Previous wood, with a high air content, has been re-
authors have suggested using wide window ported to mimic a gas collection [5]. The atten-
widths up to 1,000 H with a level of 500 H for uation values for smaller objects may vary
optimization of detection [5]. The attenuation relating to partial volume averaging. When
of a retained wooden foreign body varies in re- compared with MR imaging, CT has the ad-
lation to the content of air and fluid in the inter- vantage of being less expensive, more readily
stices of the wood. When dry wood enters the available, and faster to perform [7].
body, it is predominantly filled with air. Within The identification of wooden foreign bodies
approximately 1 week, the wood absorbs sur- may be exceedingly difficult on MR imaging,
rounding blood products and exudate and in- especially when foreign bodies are small and
creases its attenuation [6]. In addition, different there is no associated abscess or fluid collec-
types of wood have been shown to have vari- tion. In such cases, the foreign body may ap-
able attenuations, and surface coating such as pear as a signal void with surrounding
paint or sealant will affect the degree and tim- nonspecific granulation tissue. In this series, all
ing of this increase in attenuation [7]. Dry foreign bodies appeared hypointense on all se-

AJR:178, March 2002 559

 Peterson et al.
Downloaded from www.ajronline.org by 103.47.135.13 on 06/20/16 from IP address 103.47.135.13. Copyright ARRS. For personal use only; all rights reserved

A B

Fig. 3.11-year-old girl who presented for 2 years with intermit-

tently draining sinus on dorsum of foot at site of previous surgery
for ganglion.
A, Sagittal short tau inversion recovery MR image shows hy-
pointense retained wooden foreign body with surrounding high-
signal inflammatory response.
B, Three-dimensional surface rendered CT scan of foot shows
retained wooden foreign body within soft tissue between first
and second metatarsals.
C, Gross specimen photograph depicts toothpick removed at
surgery.
C

quences. Two cases showed the lesions to be
signal voids. Retained wood, in contrast to
metal, does not reveal susceptibility artifact,
and linear signal voids may be mistaken for
tendons or dense collagenous structures (Fig.
2). It has been reported with MR imaging, just
as with CT, that wood in soft tissue may absorb
the surrounding hematoma and exudate, pro-
longing T1 and T2 relaxation times [1].
Some degree of surrounding inflammatory
tissue is usually associated with a foreign
body. In the acute setting, surrounding hem-
orrhage and hematoma may be seen, being
replaced in time with granulomatous tissue.
In this study, the inflammatory reaction asso-
ciated with retained wood showed prolonged
T1 and T2 relaxation times and prominent
contrast enhancement. Identification of the
inflammatory response can assist the viewer
in identifying the retained foreign body be-
cause the actual splinter may be difficult to
visualize. The surrounding foreign body re-
action may be mistaken for a soft-tissue
mass or a tumor if the central foreign body is
not identified [8].
Sonography has been well studied in the
evaluation of retained foreign bodies and has
proved both sensitive and specific [9, 10].
Given the markedly different acoustic im-
pedance of wood and soft tissues, retained
wooden foreign bodies are easily identified,
with the leading edge of the echogenic wood
resulting in marked acoustic shadowing [11]
(Figs. 4 and 5). Sonography proved to be the
best modality in the detection of retained
wooden foreign bodies. However, the evalua-
tion was often performed to confirm findings
first seen on other modalities. In our experi-
ence, only 25% of patients presented with a
history of penetrating injury. In patients pre-
senting with nonspecific pain and swelling,
MR imaging or CT is often performed first to

560 AJR:178, March 2002

 Wooden Foreign Bodies
Downloaded from www.ajronline.org by 103.47.135.13 on 06/20/16 from IP address 103.47.135.13. Copyright ARRS. For personal use only; all rights reserved

A B

C D

Fig. 4.Retained wooden foreign body with abscess and draining sinus tract in 49-year-old man who fell off roof into foliage several months earlier.
A, Spin-echo T2-weighted axial MR image (TR/TE, 2,716/80) of right thigh reveals retained wooden foreign body (asterisk) in right vastus lateralis muscle. Hypointense
retained wood is seen centrally with surrounding hyperintense inflammatory response.
B, Sagittal short tau inversion recovery MR image (2,300/30; inversion time, 150 msec) shows foreign body as signal void with surrounding hyperintense granulomatous
response. Note associated cellulitis and sinus tract extending through subcutaneous adipose tissue.
C, Gray-scale sonogram shows hyperechoic retained wooden foreign body with associated acoustic shadowing.
D, Gross specimen photograph shows large twig removed at surgery.

evaluate for the presence of an underlying
mass or inflammatory process. At our institu-
tion, if there is any reason to suspect a re-
tained foreign body or if an equivocal case
imaged with another modality presents,
sonography is routinely performed. When
compared with MR imaging or CT, sonogra-
phy is less expensive, more readily available,
and superior in the detection of small
wooden foreign bodies. Sonography is the
modality of choice in patients who present
with a history of antecedent skin puncture or
when a penetrating injury is suspected.
Arthrography may be helpful in cases in
which the wooden foreign body has pene-
trated a joint cavity. In one patient in our se-
ries, the retained wooden foreign body
penetrated the elbow joint. Distention of the
elbow joint with contrast material showed
marked irregularity of the synovium consis-
tent with reactive synovitis (Fig. 1), which
was also seen on contrast-enhanced MR im-
aging. If intraarticular, the foreign body may
be identified on arthrography as a filling de-
fect in the contrast-filled joint.
When a soft-tissue mass is seen and there
is a possibility of a retained foreign body, the
shape of the lesion can be helpful. On cross-
sectional imaging, retained wooden foreign
bodies tend to be cylindric with a long length
and minimal width. Imaging perpendicular
to the foreign body often reveals a target ap-
pearance with the central foreign body ap-
pearing as a signal void or hypointense area
in contrast to the surrounding hyperintense
inflammatory tissue [3] (Figs. 2 and 4). If a
wooden splinter is large and the slice thick-
ness is thin, the lesion can be seen in a profile
that allows a confident diagnosis (Fig. 3).
Unfortunately, imaging parallel to a thin for-
eign body with thick slices or a large gap
may miss the wood completely and render
the foreign body inconspicuous.
In summary, the detection of retained
wooden foreign bodies can be exceedingly dif-

AJR:178, March 2002 561


Downloaded from www.ajronline.org by 103.47.135.13 on 06/20/16 from IP address 103.47.135.13. Copyright ARRS. For personal use only; all rights reserved
Fig. 5.40-year-old woman with draining sinus tract on plantar aspect of foot with multiple prior dbridements with history of penetrating trauma 2 years earlier.
A, Sagittal reformatted CT scan of forefoot shows hyperdense retained wooden foreign body in plantar soft tissues of foot.
B, Sonogram of foot shows retained wooden foreign body with hyperechoic leading edge and obscuration of trailing edge by acoustic shadowing.
ficult because patients typically present with
nonspecific symptoms without a reported his-
tory of penetrating injury. Radiographs are fre-
quently the first study obtained, but they are
usually unremarkable. MR imaging and CT
are often performed to evaluate nonspecific
symptoms; however, accurate identification of
retained wood with these modalities can be
difficult. Wood usually shows a linear hy-
pointense signal on MR imaging with an asso-
ciated inflammatory mass. CT typically shows
the retained wood as a linear area of increased
attenuation, which is best seen on wide win-
dow settings. Sonography has proved the most
useful modality, easily identifying the retained
wood as a linear echogenic focus with marked
acoustic shadowing. Unfortunately, sonogra-
phy is often underused if the appropriate his-
tory is not presented. At our institution, if there
is any reason to suspect a retained wooden
foreign body, sonography is routinely per-
562
Peterson et al.
A B
formed. Radiologists need to be aware of the
various imaging appearances of wooden for-
eign bodies and should be able to successfully
localize the retained wood if the appropriate
history is obtained.
References
1. Ochiai H, Yamakawa Y, Fukushima T, Yamada H.
Neuroimaging of a wooden foreign body retained
for 5 months in the temporalis muscle following
penetrating trauma with a chopstick. Neurol Med
Chir 1998;39:744747
2. Anderson MA, Newmeyer WL, Kilgore ES. Di-
agnosis and treatment of retained foreign bodies
in the hand. Am J Surg 1982;144:6367
3. Monu JU, McManus CM, Ward WG, Haygood TM,
Pope TL, Bohrer SP. Soft-tissue masses caused by
long-standing foreign bodies in the extremities: MR
imaging findings. AJR 1995;165:395397
4. Laor T, Barnewolt CE. Nonradiopaque penetrat-
ing foreign body: a sticky situation. Pediatr Ra-
diol 1999;29:702704
5. Ho VT, McGuckin JF, Smergel EM. Intraorbital
wooden foreign body: CT and MR appearance.
AJNR 1996;17:134136
6. Ginsberg LE, Williams DW, Mathew VP. CT in pene-
trating craniocervical injury by wooden foreign bod-
ies: reminder of a pitfall. AJNR 1993;14:892895
7. McGuckin JF, Akhtar N, Ho VT, Smergel EM,
Kubacki EJ, Villafana T. CT and MR evaluation
of a wooden foreign body in an in vitro model of
the orbit. AJNR 1996;17:129133
8. Ferguson PC, Bell RS, Davis AM. Foreign-body
abscesses presenting as soft tissue tumors: two
case reports. Can J Surg 1994;37:503507
9. Mizel MS, Steinmetz ND, Trepman E. Detection
of wooden foreign bodies in muscle tissue: exper-
imental comparison of computed tomography,
magnetic resonance imaging and ultrasonogra-
phy. Foot Ankle Int 1994;15:437443
10. Horton LK, Jacobson JA, Powell A, Fessell DP,
Hayes CW. Sonography and radiography of soft-
tissue foreign bodies. AJR 2001;175:11551159
11. Fornage BD, Schernberg FL. Sonographic diag-
nosis of foreign bodies of the distal extremities.
AJR 1986;147:567569
AJR:178, March 2002