Introduction
Soft tissue masses of the foot and ankle are relatively infrequent and may pose a diagnostic dilemma [
1]. Ultrasound (US) evaluation of soft tissue lesions is useful in the initial triaging of soft tissue lesions. US can demonstrate the cystic nature of some benign conditions such as ganglions or synovial cysts and may also be diagnostic in some other benign lesions such as superficial lipoma (if stable over 6 months), Morton neuroma, foreign body granuloma and plantar fibromatosis [
2]. However, it should be used with caution as it may be difficult to interpret and can be associated with a delay in diagnosis of malignant lesions [
3]. Magnetic resonance imaging (MRI) evaluation should follow on from US in any case where there is a reasonable chance of malignancy, where a lesion is incompletely evaluated, and in any lesion which is > 5 cm, crosses or lies deep to the superficial fascia or occurs at a site of previous resection [
2].
MRI provides excellent anatomical detail and allows for soft tissue characterisation which plays an important role in formulating a differential diagnosis. It allows for local staging and description of the relationship of a lesion to adjacent anatomical structures (e.g. fascia, bone, muscle, neurovascular structures). Furthermore, advanced MRI techniques may aid in soft tissue characterisation. Proton magnetic resonance spectroscopy (PMRS) may demonstrate an elevated choline peak within a soft tissue lesion, indicative of the high cellular turnover seen in malignancy [
4,
5]. Susceptibility weighted imaging (SWI) allows for easier detection of low signal haemosiderin, melanin and calcification within a lesion, as found for example in metastatic melanoma [
4]. Lower apparent diffusion coefficient (ADC) scores on diffusion-weighted imaging (DWI) are generally seen in malignant lesions compared with benign lesions. ADC values may also help in differentiating necrotic malignant lesions from abscesses. Greater diffusion restriction is seen within the necrotic core of an abscess compared with necrotic tumours, which tend to have greater diffusion restriction in the solid tumour components [
4,
5]. Myxoid tumours have been shown to have a higher ADC value than non-myxoid lesions [
4‐
6]. Dynamic contrast-enhanced (DCE) perfusion imaging may aid in differentiating benign from malignant lesions based on patterns of contrast enhancement and washout. In addition, it can assess for viable, enhancing tumour to guide biopsy and may aid in early detection of residual/recurrent tumour post resection [
4,
5]. It should be kept in mind that considerable overlap between benign and malignant lesions has been found with all of these techniques. Radiomic analysis of MR images has been shown to have some prognostic value in soft tissue sarcoma but is not yet a mature technique [
7,
8].
Plain radiography is almost always useful as an adjunct to US. It allows characterisation of mineralisation within a lesion and can be useful to assess for the presence of phleboliths. It is also useful to assess the extent of underlying bone and joint involvement, which can both suggest an aetiology and have important implications for surgical treatment. Computed tomography (CT) can be very helpful in the foot and ankle because of the complex anatomy and overlapping structures.
Fluorine-deoxyglucose positron emission tomography/computed tomography (FDG PET/CT) is not usually indicated in the initial evaluation of a soft tissue lesion. There is however a correlation between standardised uptake value (SUV) max and histologic subtype in soft tissue sarcomas with high grade tumours demonstrating a higher SUV max value. It has also been shown to be very sensitive in assessing for distant metastatic disease, in directing a target for biopsy (avoiding non-diagnostic samples secondary to tumour necrosis) and in evaluating treatment response [
9,
10].
A variety of neoplastic and non-neoplastic lesions may occur around the foot and ankle, the vast majority of which are benign. The most common malignant lesions are synovial sarcoma and pleomorphic undifferentiated sarcoma [
11]. The most common benign lesions include plantar fibromatosis, lipoma, peripheral nerve sheath tumours and giant cell tumours of the tendon sheath [
12]. Table
1 shows the breakdown of the most common lesions by age group. Malignant neoplastic processes can mimic benign lesions and may be difficult to differentiate [
3,
13]. This highlights the importance of a thorough knowledge of the imaging characteristics of these lesions. The non-neoplastic “pseudotumoural” lesions have been covered in this journal previously by Van Hul et al. [
13]. This review focuses on the MRI characteristics of the most common neoplastic lesions of the foot and ankle, both benign and malignant.
Table 1
Table showing the most common benign and malignant lesions of the foot and ankle subdivided by age group as described by Kransdorf [
7,
8]
0–5 years | • Granuloma annulare • Fibromatosis • Haemangioma | 30 25 11 | • Fibrosarcoma • DFSP • MPNST • Rhabdomyosarcoma | 45 18 18 18 |
6–15 years | • Fibromatosis • Granuloma annulare • Haemangioma | 23 13 13 | • Synovial sarcoma • DFSP • Rhabdomyosarcoma | 21 17 9 |
16–25 years | • Fibromatosis • GCTTS • Granuloma annulare • Fibrous histiocytoma | 22 14 12 12 | • Synovial sarcoma • Clear cell sarcoma • Fibrosarcoma • DFSP | 30 11 8 8 |
26–45 years | • Fibromatosis • Fibrous histiocytoma • PNST • GCCTS | 25 13 11 9 | • Synovial sarcoma • Clear cell sarcoma • Pleomorphic undifferentiated sarcoma | 26 13 13 |
46–65 years | • Fibromatosis • Fibrous histiocytoma • Lipoma • PNST | 25 13 11 8 | • Pleomorphic undifferentiated sarcoma • Synovial sarcoma • Leiomyosarcoma • Kaposi’s sarcoma | 25 17 12 9 |
66 and over | • Fibromatosis • PNST • Fibrous histiocytoma • Chondroma | 14 13 11 9 | • Kaposi’s sarcoma • Pleomorphic undifferentiated sarcoma • Leiomyosarcoma | 37 19 15 |
Summary
MRI is the imaging modality of choice when dealing with soft tissue lesions of the foot or ankle. Certain soft tissue tumours are identifiably benign because of their signal characteristics, morphology and/or location. These include plantar fibromatosis, haemangioma, lipoma, PVNS/GCT tendon sheath and synovial chondromatosis. Sometimes, soft tissue chondroma and peripheral nerve sheath tumours will be relatively characteristic in appearance. For other non-specific soft tissue masses, a high level of suspicion for malignancy and a low threshold for obtaining tissue should be maintained.
A lesion lacking features of a specific benign entity should be regarded as malignant until proven otherwise. Features that are more common in malignant than in benign lesions include large size (> 5 cm), deep site, inhomogeneous signal intensity, haemorrhage and necrosis, early and inhomogeneous contrast enhancement, irregular margins, surrounding soft tissue oedema and invasion of adjacent structures, including bone and neurovascular structures [
17]. Advanced MRI techniques such as spectroscopy, perfusion and diffusion-weighted imaging may contribute to better soft tissue characterisation.
Referral for biopsy is indicated when there is a reasonable suspicion of malignancy, in any superficial lesion > 5 cm, any deep lesion regardless of size and in indeterminate cases. Biopsy should be planned and performed in a regional sarcoma referral centre, involving multidisciplinary input, including that of an oncologic surgeon. Core biopsy is required as fine needle aspiration is inadequate. The approach should be planned to avoid traversing non-affected compartments and the biopsy tract removed en bloc at the time of surgery. MRI, US and FDG PET may be used to guide biopsy to sites of viable tumour, avoiding sites of necrosis or haemorrhage.
A comprehensive knowledge of the imaging characteristics of the soft tissue lesions which occur around the foot and ankle will enable a radiologist to confidently identify those lesions which have a characteristic imaging appearance and, most importantly, to identify those lesions which require further imaging or surgical referral.
Funding
The authors, their immediate family and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article. There was no outside funding or grants received that assisted in this study, and no outside source of funds was involved in data collection, data analysis or the preparation of or editing of the manuscript.