The Egyptian Journal of Radiology and Nuclear Medicine (2013) 44, 573–579

Egyptian Society of Radiology and Nuclear Medicine

The Egyptian Journal of Radiology andNuclearMedicine

www.elsevier.com/locate/ejrnmwww.sciencedirect.com

ORIGINAL ARTICLE

Wooden splinter-induced extremity injuries: Accuracy

of MRI evaluation

Mohamed Ragab Nouh a,d,*,1, Ahmed Mohamed Sabry Nasr b,2,

Mohamed Osama El-Shebeny c,3

a Department of Radiology, Faculty of Medicine, Alexandria University, Alexandria, Egyptb Department of Radiology, Faculty of Medicine, Zagazig University, Zagazig, Egyptc Department of General Surgery, Damanhour Teaching Hospital, Damanhour, Egyptd Department of Radiology, Al-Razi Orthopedic Hospital, Sulibikhate, Kuwait

Received 8 January 2013; accepted 3 June 2013

Available online 27 June 2013

*

Fa

Te

E-

(M

m1

A2

Te3

Pe

N

03

ht

KEYWORDS

MRI;

Wooden splinters;

Extremity swellings

Corresponding author. Add

culty of Medicine, Alexandri

l.: +20 65099562/24825909;

mail addresses: mragab73

.R. Nouh), anasr1998@yah

ail.com (M.O. El-Shebeny).

Department of Radiology,

bd El-Naser Street, Sulibikha

Department of Radiology,

l.: +965 226380591.

Tel.: +20 453318222.

er review under responsibility

uclear Medicine.

Production an

78-603X � 2013 Production

tp://dx.doi.org/10.1016/j.ejrn

ress: De

a Univer

fax: +20

@yahoo.

oo.com (

Al-Razi

te 13001

Chest H

of Egyp

d hostin

and host

m.2013.0

Abstract Objective: To detect the accuracy of MR imaging in detection and localization of woo-

den splinters invading the extremities using surgical data as a reference standard.

Methods: A retrospective review on a series of eighteen patients with: history of wooden foreign

body penetration and/or localized swellings to their extremities, surgically confirmed final diagnosis

of wooden foreign body penetration and having both screening X-ray and MR imaging of their

concerned extremities.

MR imaging included variable combination of fast-spin echo imaging in T1W and T2W without

fat-suppression as well as fat-suppressed proton density and/or STIR sequences. Gadolinium-

enhanced imaging was available in 10 of the MR studies of our patients only.

Results: Successful localization using MR was achieved in sixteen patients only, in the current

study with sensitivity and specificity of 88.8%. Wooden splinters were recognized as linear signal

partment of Radiodiagnosis,

sity, Alexandria 21599, Egypt.

24825508.

com, mragab73@gmail.com

A.M.S. Nasr), oelshebeny@g-

Orthopedic Hospital, Gamal

, Kuwait.

ospital, Sulibikhate, Kuwait.

tian Society of Radiology and

g by Elsevier

ing by Elsevier B.V. on behalf of Egyptian Society of Radiology and Nuclear Medicine.

6.001

574 M.R. Nouh et al.

void on T1, T2 and PD weighted images surrounded by reactive inflammatory changes. In the

remaining two cases, wooden splinters were missed on MRI due to their tiny size.

Conclusions: Characteristic target appearance of signal void wooden splinter and surrounding

inflammatory changes on MR imaging of unusual soft-tissue masses of the extremities should alert

radiologist to the possibility of wooden-splinter induced granuloma.

� 2013 Production and hosting by Elsevier B.V. on behalf of Egyptian Society of Radiology and Nuclear

Medicine.

1. Introduction

Most foreign body (FB) injuries to the extremities commonlyoccur during daily activities and in the context of occupational

exposures (1). Patients usually present with non-specific symp-toms such as pain and/or swelling and it will be a predicamentif no previous history of trauma is recognized, making its

detection important to avoid serious complications (2,3).Plain X-ray screening to exclude foreign bodies after trau-

ma is a widely used practice (2,3). Yet, it is unrewarding for

non-radiopaque foreign bodies such as wood that frequentlyremain undetected and represent a diagnostic challenge, espe-cially if foreign body penetration history is uncertain (2,4).

Ultrasound (US) provided an excellent alternative method

for identifying, localizing and even retrieval of radiolucent for-eign bodies (4–6). Some researchers emphasized the role ofcomputed tomographic (CT) scanning and magnetic resonance

imaging (MRI) in the detection of many foreign bodies thatmay be missed on radiographs (7–9). Few reports addressedthe role of MRI in the evaluation of wooden foreign bodies

in human extremities (10–12) while the major bulk of literaturedescriptions are based on case studies.

The purpose of this study was to detect the accuracy of MR

imaging in detection and localization of wooden splintersembedded in the extremities using surgical data as a referencestandard.

2. Materials and methods

Following our medical ethics committee approval exemptingindividual informed consent acquirement, we carried out a ret-

rospective study on a series of patients presented to our insti-tution; over a period of 12 months (between 1st April 2010 and31st March 2011); using our institutional data and minor pro-

cedure registries.Our study population included 18 patients (16 males and 2

females) achieving the following inclusion criteria:

(a) Patients with suspected wooden foreign body penetra-tion presented with painful extremities with or without

localized swellings.(b) Patients had plain X-rays that were negative for radio-

opaque foreign bodies.(c) Patients underwent MR imaging in their diagnostic

work up to search possible non- radiopaque foreignbodies.

(d) Patients with surgically confirmed final diagnosis of

wooden foreign body penetration.

Patients with surgically confirmed non-wooden foreign

bodies e.g. glass pieces and stones were excluded from ourstudy population.

Patient’s age, sex, occupation and accidental circumstancesassociated with foreign body injury were recorded to evaluatepatient’s demography.

All patients had X-ray screening of their concerned extrem-ities at initial presentation. MR imaging evaluation of all pa-tients was carried out on 1.5T units. In our institution,

extremity imaging is carried out using a suitable size flexiblecoil or a quadrature send-receive knee coil. The patient isimmobilized in a comfortable position to limit possible exhaus-tion-induced motion artifacts. The imaging series usually in-

cludes variable combination of fast-spin echo imaging inT1W spin-echo (SE) (TR/TE = 500/12, NEX = 2) and T2Wfast spin-echo (FSE) (TR/TE = 4500/80, NEX = 3) without

fat-suppression in sagittal, short-axis (coronal) and long axis(axial) planes of the foot. Additional fat-suppressed protondensity (TR/TE = 3000/35) and STIR (TR/TE = 5400/45,

TI = 140) sequences may be acquired in best planes to opti-mize the concerned anatomy and allow good delineation ofthe pathologic findings. Gadolinium-enhanced imaging wasavailable in 10 of the MR studies of our patients.

All radiographic and MR imaging studies of our patient’spopulation were reviewed by the authors in consensus.

The radiographic studies were assessed for the presence of

bony changes including cortical fractures and periosteal reac-tion as well as soft-tissue changes e.g. increased densities, pres-ence of gas and/or swellings.

The MR images were evaluated for the presence of a signalvoid foreign body, soft-tissue edema and/or collection, massand/or abscess formation as well as the presence of any adja-

cent bony changes or articular abnormalities.Twelve patients were found to have had targeted US exam-

inations at outside facilities prior to presenting to our institu-tion. Studies were performed to evaluate their localized soft

tissue swelling and/or site of suspected foreign body entry.None of our cases underwent CT or invasive procedures suchas sinography.

3. Results

All relevant patients’ data were processed and analyzed as re-

gards their demographic, clinical and imaging perspectives.

3.1. Demographic and clinical data

Our study population included 18 patients, two (11.1%) fe-males and sixteen (88.8%) males with an age range of 3–50 years (mean age, 23.94 years).

Seven (38.8%) of our patients were uncertain of previoustraumatic history while the remaining eleven (61.1%) patientsgave a previous history of antecedent skin puncture and/or pe-netrating injury. In this latter group, the time interval between

inciting injury and presenting symptoms varied between

Fig. 1 (a and b): a 3 year old boy fell on a palm stent while

playing and a palm thorn was then removed from his forearm by

his father who subsequently took him to the emergency depart-

ment. Initial radiograph showed no abnormalities. (a) Three weeks

later, the child presented with a painful swollen forearm along

with low grade fever. Follow up radiograph at the time revealed

alternating areas of bone lucencies and sclerosis of the ulnar shaft

associated with prominent periosteal reaction consistent with

osteomyelitis.

Wooden splinter-induced extremity injuries: Accuracy of MRI evaluation 575

14 days and 8 months with a mean average interval of68.5 days.

Clinically, localized extremity swelling was the commonest

presenting symptom in the current work (Table 1). Anatomi-cally; sole of the foot was the commonest target for woodensplinter among our study population, (Table 2).

3.2. Radiographic examinations

Soft-tissue swelling accompanied by osteomyelitis of the ulna

was seen in a 3 year old child (Fig. 1). Subtle soft tissue fullnessof the dorsum of the foot was noted in another case. Theremaining sixteen cases had unremarkable radiographic stud-

ies. None of the wooden splinters was detected on X-ray inour series.

3.3. MRI examinations

Sixteen of our patients revealed the wooden splinters as signalvoid structure on T1, T2 and PD weighted images surroundedby reactive inflammatory changes manifest as low signal on

T1W images and high signal on T2W and PDW images. Fiveof them showed the wooden splinters in a leading sinus tractconnecting it to the adjacent superficial skin; either in foot

or hand. The remaining eleven cases showed variable inflam-matory reactive changes including: abscess formation(Fig. 2); muscle edema (Fig. 3), and indurated inflammatorymass (Fig. 4); within the involved extremity.

All 10 cases in which intravenous gadolinium administra-tion was utilized demonstrated peripheral enhancement ofthe inflammatory reaction surrounding the wooden splinters.

Central liquefaction and abscess formation surrounding theembedded FB were noted in two cases. Mild enhancement ofcalcaneal marrow, believed to be due to reactive marrow ede-

ma was seen in one case (Fig. 3). Osteomyelitis of the ulna withextensive marrow enhancement and associated contiguous softtissue abscess was noted in another (Fig. 2). Successful locali-

zation could be achieved in only sixteen patients (Table 3). Thefoot quadrates plantae and/or flexor hallucis brevis were themost common location for wooden splinter.

Table 1 Presenting clinical complaints of the study

population.

Presenting complaint Number of patients

Painful extremity abscess 3 (16.6%)

Localized extremity mass 14 (77.7%)

Charging skin sinus 1 (5.5%)

Total number 18 (100%)

Table 2 Entry site and wooden splinter dormancy distribu-

tion of study population.

Wooden splinter location Number of patients

Fore-arm 1(5.5%)

Palm of the hand 2(11.1%)

Foot sole 13(72.2%)

Foot dorsum 2(11.1%)

Total number 18(100%)

In one of the foot cases MR failed to detect a tiny (3 mm)wooden splinter which was subsequently retrieved surgically.

The splinter was found embedded within granulomatoussoft-tissue mass involving the oblique head of abductor hallu-cis muscle, adjacent interossi and plantar lumbericals (Fig. 5).In another case, the small wooden splinter was partly embed-

ded in the ulnar cortex and was incorrectly interpreted as sig-nal void pertaining to an engorged nutrient vessel (Fig. 2).Onsurgical exploration evacuation of soft-tissue abscess and

debridement revealed a small (3 mm) wooden splinter to bepartly inserted within the postero-lateral ulnar cortex thatpassed undetected on MRI.

Using surgical retrieval of the wooden splinter as the goldstandard, both the sensitivity and specificity of MR for thedetection of wooden splinters were approximately 88.8%.

3.4. Additional diagnostic imaging work-up

Twelve of our patients underwent targeted ultrasound at out-side facilities prior to being admitted to our institution. Seven

of the twelve patients showed parts of the wooden foreignbodies in a relatively superficial location where as, in the

Fig. 2 (a through e): same child as in Fig. 1; axial fat-saturated PDW (a) and non-contrast T1 SE (b) show ulnar shaft marrow signal

abnormality with extensive surrounding periosteal reaction, edema and subperiosteal abscess manifested with high-signal proteinaceous

content. The contrast enhanced T1 SE axial (c), sagittal (d) and coronal (e) planes demonstrate extensive marrow and surrounding soft

tissue enhancement with a non-enhancing fluid collection dorsally consistent with an abscess. The small signal void structure (black arrow)

seen along the radial aspect of the ulna was mis-interpreted as a nutrient vessel but was later determined to be a small wooden splinter

upon surgery.

576 M.R. Nouh et al.

remaining five patients, ultrasound could not demonstrate theforeign body necessitating further MR imaging evaluation. In

the remaining six cases ultrasound examinations were notfound in their records.

None of our cases underwent CT evaluation and/or inva-

sive imaging procedures as sinography.

4. Discussion

Wooden splinters and thorn injuries of the extremities rep-resent common emergency department admissions (2,3).Upon cutaneous penetration, they usually cause prominent

regional inflammation about the entry site, form a leadingtract or become engulfed amidst inflammatory reactionthus, forming a granuloma in the neglected state (13). Olderinjuries, with forgotten and/or unrecognized history of for-

eign-body exposure, may present as infection, inflammation,induration, or mass formation representing a sticky clinicalscenario (10–15).

Our series revealed middle age and male preponderance forextremity foreign body injuries. This could be explained by thepredominant involvement of young males in manual labor,

particularly in developing countries. Also, the median intervalof about 68.5 days between the inciting injury and onset of pre-

senting symptoms which was encountered in our study is sim-ilar to the long durations that were reported in the literature(16,17).

Our study revealed that foot is commonly subject towooden splinter injuries more than hand. This is contradic-tory to previous reports emphasizing that foot is the secondcommonest location for retained foreign bodies following

hand (14,16,18). However, this contradiction could be attrib-uted to accidental circumstances, socio-economic status andtraditional habits as barefoot during agriculture work in

developing countries.Radiography has showed a sensitivity of <15% in

detecting wooden foreign bodies in hand (3) thanks to the

porous nature as well as air within and surrounding it or dueto adjacent bony changes that may accompany its retention(2). This figure is lowered to about 5% in a recent study, by

Levine et al. (15).In our series, X-ray failed to visualize the retained woo-

den splinter in any of our patients emphasizing the difficultyof detection of wooden foreign bodies on radiography (2–

Fig. 3 (a through d): an 11 year old boy presented with painful heel following a 1 week old tree stent prick while playing football. Non-

contrast sagittal T1W SE (a), fat-saturated short-axis PD (b), as well as, post-contrast T1W SE in short axis (c) and sagittal plane (d) show

a linear signal void wooden splinter penetrating the proximal medial bundle of the plantar plate and creating an enhancing sinus tract

within the subcutaneous tissues, quadratus plantae and flexor digitorum brevis muscles. Note minimal intermuscular fluid collections in

(b), enhancing reactive marrow edema on (c). No osteomyelitis was found upon surgery or during the post-operative period.

Fig. 4 (a through d): a 33 year old male carpenter presented with 8 month long duration painful swelling of his foot dorsum. Short-axis

fat-saturated PD (a), short-axis T1W SE (b), post-contrast T1W SE in short-axis (c) and sagittal (d) planes showing the signal void

wooden splinter penetrating the dorsal subcutaneous tissues over the first dorsal inter-metatarsal space. Note the enhancing phlegmon

with the wooden splinter inside. No dorsal inter-osseous affection was present.

Wooden splinter-induced extremity injuries: Accuracy of MRI evaluation 577

4,11–13). In only one case in our series, bony changes con-sistent with osteomyelitis of the ulna of a young child were

radiographically apparent.

US has been proved as an excellent alternative method foridentifying, localizing and even retrieval of superficial radiolu-

cent foreign bodies (5,6). However, it is limited by its operator

Table 3 Distribution of wooden splinters location using MR

among study population*.

Site of wooden splinter Number

of cases

Superficial to palmar aponeurosis 1

Deep to palmar aponeurosis 1

Foot extensor digitorum and dorsal lumbericals 2

Foot quadrate plantae and/or flexor hallucis brevis 7

Foot abductor hallucis 2

Foot flexor digiti minimi 2

Foot deep planter interossei 1

Total 16

* Two cases were missed on MR studies and wooden splinter

invasion was diagnosed on surgery.

578 M.R. Nouh et al.

dependence, and inadequate localization of deeply seated for-eign body. Likewise, we encountered similar results in those

patients who had undergone US examinations in our series.Despite the advancement in imaging modalities, the detec-

tion of deeply seated foreign bodies in the extremities remains

a challenging task. Wood predominately enters the body dryand with full of air pockets and soon absorbs the resultant sur-rounding inflammatory exudates rendering MRI a valuable

tool for its evaluation, especially in the extremities(11,16,17,19–21).

Fig. 5 (a through c): a 16 year old male farmer presented with 10

contrast short-axis PD (a), post-contrast T1W SE in long axis (b); and

tumefaction of the flexor digitorum brevis with no neo-matrix format

(2 mm) wooden foreign body within inflamed muscle was retrieved. A

operative period.

In our series, MR revealed the wooden splinters as signalvoid structures on T1W, T2W and PDW images even in theabsence of associated abscess. It was longitudinal in shape

when scans are acquired parallel to it and dot like when scansare oriented perpendicular to it. The sensitivity and specificityof our current study were about 88.8%. In the two cases that

were missed by MR, the wooden splinters were tiny in size(<3 mm) and were subsequently retrieved surgically.

These inflammatory reactions appeared as a ring of T1W

hypo-intensity, T2W and PDW hyper-intensity and variabledegrees of enhancement following intra-venous administra-tion of gadolinium surrounding the signal void woodensplinter. This resulted in characteristic ‘‘target-like’’ appear-

ance when imaging planes are perpendicular to their dor-mancy. This compares well to previous literaturedescriptions of wooden foreign body MR appearance

appearing as linear hypointense signal on T1W and T2WMR sequences especially if associated with inflammatorymass (8–12,22).

Some recommended use of T1-weighted images as it maxi-mize contrast between fat and wood (23) while others statedthat high signal intensity of the granulation tissue may out-

shine the foreign body and may be a cause for its difficult iden-tification on T2-weighted images (16). We did not find thatsignificant in our series.

Our study is limited by its retrospective nature and small

number of our study population. While a timely performed

months of sole pain and failed to recall history of trauma. Non-

sagittal (c) planes of the foot showing a deeply seated enhancing

ion or surrounding tenosynovitis. On surgical exploration, a tiny

ntibiotics were given and patient recovered well during the post-

Wooden splinter-induced extremity injuries: Accuracy of MRI evaluation 579

MRI is an expensive; time consuming modality, it remains rel-atively cost effective given the potential complications andeventual expenses of the evaluation of extremity soft tissue

masses related to unrecognized embedded radiolucent foreignbody such as wooden splinters.

In conclusion, retained wooden splinters should be consid-

ered on interpretation of negative radiograms in the setting ofpenetrating injuries to extremities. Also, soft tissue changes re-lated to unrecognized embedded wooden splinters have to be

considered upon imaging unusual soft tissue masses of theextremities. This is particularly true for patients with propen-sity for this type of injury even with an inconclusive history.Finally, the ‘target-like’ appearance depicting the signal void

wooden splinter with surrounding inflammatory stigmata onvarying imaging sequences is characteristic.

Disclosures

No financial or conflict of interest to declare.

References

(1) Hunter TimB, Taljanovic MihraS. Foreign Bodies. RadioGraph-

ics 2003;23:731–57.

(2) Laor T, Barnewolt CE. Non-radiopaque penetrating foreign

body: ‘a sticky situation’. Pediatr Radiol 1999;29:702–4.

(3) Anderson MA, Newmeyer III WL, Kilgore Jr ES. Diagnosis and

treatment of retained foreign bodies in the hand. Am J Surg

1982;144:63–7.

(4) Horton KL, Jacobson JA, Powell A, Fessell DP, Hayes CW.

Sonography and radiography of soft-tissue foreign bodies. AJR

2001;176:1155–9.

(5) Jacobson JA, Powell A, Craig JG, Bouffard JA, Van Holsbeeck

MT. Wooden foreign bodies in soft tissue: detection at US.

Radiology 1998;206:45–8.

(6) Shiels II WE, Babcock DS, Wilson JL, Burch RA. Localization

and guided removal of soft-tissue foreign bodies with sonography.

AJR 1990;155:1277–81.

(7) McGuckin Jr 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;17:129–33.

(8) Mizel MS, Steinmetz ND, Trepman E. Detection of wooden

foreign bodies in muscle tissue: experimental comparison of

computed tomography, magnetic resonance imaging, and ultra-

sonography. Foot Ankle Int 1994;15:437–43.

(9) Ober CP, Jones JC, Larson MM, Lanz OI, Were SR. Comparison

of ultrasound, computed tomography, and magnetic resonance

imaging in detection of acute wooden foreign bodies in the canine

manus. Vet Radiol Ultrasound 2008;49(5):411–8.

(10) Bodne D, Quinn SF, Cochran CF. Imaging foreign glass and

wooden bodies of the extremities with CT and MRI. J Comput

Assist Tomogr 1988;12:608–11.

(11) Peterson JJ, Bancroft LW, Kransdorf MJ. Wooden foreign

bodies: imaging appearance. AJR 2002;178:557–62.

(12) Ando A, Hatori M, Hagiwara Y, Isefuku S, Itoi E. Imaging

features of foreign body granuloma in the lower extremities

mimicking a soft tissue neoplasm. Ups J Med Sci 2009;114:46–51.

(13) Chan C, Salam GA. Splinter Removal. Am Fam Physician

2003;7(12):2557–62.

(14) Borgia CA. An unusual bone reaction to an organic foreign body

in the hand. Clin Orthop Relat Res 1963;30:188–93.

(15) Levine MR, Gorman SM, Young CF, Courtney DM. Clinical

characteristics and management of wound foreign bodies in the

ED. Am J Emerg Med 2008;26(8):918–22.

(16) Durr HD, Stabler A, Muller PE, Refior HJ. Thorn induced

pseudo-tumor of the metatarsal. J Bone Joint Surg Am 2001;83

A(4):580–4.

(17) Bode KS, Haggerty CJ, Krause J. Latent foreign body synovitis. J

Foot Ankle Surg 2007;46(4):291–6.

(18) Reginato AJ, Ferreiro JL, O’Connor CR, Barbasan C, Arasa J,

Bednar J, et al. Clinical and pathological studies of twenty-six

patients with penetrating foreign body injury to the joints, bursae,

and tendon sheaths. Arthritis Rheum 1990;33:1753–62.

(19) Guner S, Ceylan MF, Isik D, Guner SI, Ediz L. A case of wooden

foreign body retained in the calcaneus. Pak J Med Sci

2011;27(4):932–4.

(20) Kornreich L, Katz K, Horev G, Zeharia A, Mukamel M.

Preoperative localization of a foreign body by magnetic resonance

imaging. Eur J Radiol 1998;26:309–11.

(21) Dastgir N, O’Rourke P. Pseudotumor of metatarsal: a thorny

problem. Int J Orthopedic Surg 2003;1(2).

(22) Monu JU, McManus CM, Ward WG, Haygood TM, Pope TM,

Pope TL, et al. Soft-tissue masses caused by long-standing

foreign bodies in the extremities: MR imaging findings. AJR

1995;165:395–7.

(23) Krimmel M, Cornelius CP, Stojadinovic S, Hoffmann J, Reinert

S. Wooden foreign bodies in facial injury: a radiological pitfall.

Int J Oral Maxillofac Surg 2000;29:428–9.